# LLM EU — full catalogue > Generated from the catalogue rows. Every number, licence and date is the value > the model card records, with its source. "unverified" means the publisher has not > stated it. 325 of 617 models have a live endpoint that is > not the mock backend. LLM EU is a private company, not an EU institution. ## AionLabs: Aion 3.5 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/aion-3.5-mini/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as aion-labs/aion-3.5-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## AionLabs: Aion-2.0 - Provider: OpenRouter - Page: https://llmeu.com/en/models/aion-2.0/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as aion-labs/aion-2.0. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## AionLabs: Aion-3.0-Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/aion-3.0-mini/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as aion-labs/aion-3.0-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## AionLabs: Aion-RP 1.0 (8B) - Provider: OpenRouter - Page: https://llmeu.com/en/models/aion-rp-llama-3.1-8b/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as aion-labs/aion-rp-llama-3.1-8b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Amazon: Nova 2 Lite - Provider: OpenRouter - Page: https://llmeu.com/en/models/nova-2-lite-v1/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as amazon/nova-2-lite-v1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Amazon: Nova Lite 1.0 - Provider: OpenRouter - Page: https://llmeu.com/en/models/nova-lite-v1/ - Context window: 300000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as amazon/nova-lite-v1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Amazon: Nova Micro 1.0 - Provider: OpenRouter - Page: https://llmeu.com/en/models/nova-micro-v1/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as amazon/nova-micro-v1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Amazon: Nova Pro 1.0 - Provider: OpenRouter - Page: https://llmeu.com/en/models/nova-pro-v1/ - Context window: 300000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as amazon/nova-pro-v1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Haiku 4.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-haiku-4.5/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-haiku-4.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Haiku 4.5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-haiku-4.5:batch/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-haiku-4.5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Haiku 5.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-haiku-5.5/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-haiku-5.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Haiku 5.5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-haiku-5.5:batch/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-haiku-5.5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Haiku Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-haiku-latest/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~anthropic/claude-haiku-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Sonnet 5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-sonnet-5:batch/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-sonnet-5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Anthropic: Claude Sonnet 5.5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/claude-sonnet-5.5:batch/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as anthropic/claude-sonnet-5.5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Apodex: Apodex 1.1 Mini (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/apodex-1.1-mini:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as apodex/apodex-1.1-mini:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Arcee AI: Trinity Large Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/trinity-large-thinking/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as arcee-ai/trinity-large-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed 1.6 - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-1.6/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-1.6. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed 1.6 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-1.6-flash/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-1.6-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed 2.1 Turbo - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-2-1-turbo/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-2-1-turbo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed-2.0-Code - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-2.0-code/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-2.0-code. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed-2.0-Lite - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-2.0-lite/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-2.0-lite. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance Seed: Seed-2.0-Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/seed-2.0-mini/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance-seed/seed-2.0-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ByteDance: UI-TARS 7B - Provider: OpenRouter - Page: https://llmeu.com/en/models/ui-tars-1.5-7b/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as bytedance/ui-tars-1.5-7b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Cohere: Command A+ - Provider: OpenRouter - Page: https://llmeu.com/en/models/command-a-plus/ - Context window: 192000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as cohere/command-a-plus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Cohere: Command R (08-2024) - Provider: OpenRouter - Page: https://llmeu.com/en/models/cohere-command-r-08-2024/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as cohere/command-r-08-2024. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Cohere: Command R7B (12-2024) - Provider: OpenRouter - Page: https://llmeu.com/en/models/command-r7b-12-2024/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as cohere/command-r7b-12-2024. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Cohere: North Mini Code (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/north-mini-code:free/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as cohere/north-mini-code:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek Flash Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-flash-latest/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~deepseek/deepseek-flash-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-chat/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-chat. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3 0324 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-chat-v3-0324/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-chat-v3-0324. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-chat-v3.1/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-chat-v3.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3.1 Terminus - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v3.1-terminus/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v3.1-terminus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3.2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v3.2/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v3.2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V3.2 Exp - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v3.2-exp/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v3.2-exp. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Flash 0423 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-deepseek-v4-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Flash 0731 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-deepseek-v4-flash-0731/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4-flash-0731. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Flash Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v4-flash-latest/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~deepseek/deepseek-v4-flash-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Flash Vision Exp - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v4-flash-vision-exp/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4-flash-vision-exp. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Pro 0423 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-deepseek-v4-pro/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4 Pro 0813 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v4-pro-0813/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4-pro-0813. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4.1 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v4.1-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4.1-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: DeepSeek V4.1 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-v4.1-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-v4.1-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: R1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-deepseek-r1/ - Context window: 64000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-r1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## DeepSeek: R1 0528 - Provider: OpenRouter - Page: https://llmeu.com/en/models/deepseek-deepseek-r1-0528/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as deepseek/deepseek-r1-0528. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Dots Studio: Dots3-Note Preview (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/dots-3-note-preview:free/ - Context window: 512000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as dots-studio/dots-3-note-preview:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 2.5 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-2.5-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-2.5-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 2.5 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-2.5-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-2.5-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 2.5 Flash Lite - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-2.5-flash-lite/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-2.5-flash-lite. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 2.5 Flash Lite (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-2.5-flash-lite:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-2.5-flash-lite:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 2.5 Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-2.5-pro:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-2.5-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3 Flash Preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3-flash-preview/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3-flash-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3 Flash Preview (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3-flash-preview:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3-flash-preview:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.1 Flash Lite - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.1-flash-lite/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.1-flash-lite. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.1 Flash Lite (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.1-flash-lite:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.1-flash-lite:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.1 Flash Lite Preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.1-flash-lite-preview/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.1-flash-lite-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.5 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.5-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.5-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.5 Flash Lite - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.5-flash-lite/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.5-flash-lite. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.5 Flash Lite (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.5-flash-lite:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.5-flash-lite:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.6 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.6-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.6-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.6 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.6-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.6-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.7 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.7-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.7-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.7 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.7-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.7-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.8 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.8-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.8-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini 3.8 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-3.8-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemini-3.8-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemini Flash Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemini-flash-latest/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~google/gemini-flash-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 2 27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemma-2-27b-it/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-2-27b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 3 12B - Provider: OpenRouter - Page: https://llmeu.com/en/models/google-gemma-3-12b-it/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-3-12b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 3 27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/google-gemma-3-27b-it/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-3-27b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 3 4B - Provider: OpenRouter - Page: https://llmeu.com/en/models/google-gemma-3-4b-it/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-3-4b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 4 26B A4B - Provider: OpenRouter - Page: https://llmeu.com/en/models/google-gemma-4-26b-a4b-it/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-4-26b-a4b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 4 26B A4B (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemma-4-26b-a4b-it:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-4-26b-a4b-it:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 4 31B - Provider: OpenRouter - Page: https://llmeu.com/en/models/google-gemma-4-31b-it/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-4-31b-it. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Google: Gemma 4 31B (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gemma-4-31b-it:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as google/gemma-4-31b-it:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## IBM: Granite 4.0 Micro - Provider: OpenRouter - Page: https://llmeu.com/en/models/granite-4.0-h-micro/ - Context window: 131000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ibm-granite/granite-4.0-h-micro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## IBM: Granite 4.2 8B - Provider: OpenRouter - Page: https://llmeu.com/en/models/granite-4.2-8b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ibm-granite/granite-4.2-8b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Inception: Mercury 2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mercury-2/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inception/mercury-2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Inception: Mercury 2.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mercury-2.5/ - Context window: 260000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inception/mercury-2.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Inference.net: Schematron V2 Small - Provider: OpenRouter - Page: https://llmeu.com/en/models/schematron-v2-small/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inference-net/schematron-v2-small. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Inference.net: Schematron V2 Turbo - Provider: OpenRouter - Page: https://llmeu.com/en/models/schematron-v2-turbo/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inference-net/schematron-v2-turbo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## LiquidAI: LFM2.5-2.6B (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/lfm-2.5-2.6b:free/ - Context window: 65536 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as liquid/lfm-2.5-2.6b:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mancer: Weaver (alpha) - Provider: OpenRouter - Page: https://llmeu.com/en/models/weaver/ - Context window: 8000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mancer/weaver. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meituan: LongCat 2.0 - Provider: OpenRouter - Page: https://llmeu.com/en/models/longcat-2.0/ - Context window: 1048756 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meituan/longcat-2.0. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 3.1 70B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-3.1-70b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-3.1-70b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 3.1 8B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-3.1-8b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-3.1-8b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 3.2 1B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-3.2-1b-instruct/ - Context window: 60000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-3.2-1b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 3.2 3B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-3.2-3b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-3.2-3b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 3.3 70B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-3.3-70b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-3.3-70b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 4 Maverick - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-4-maverick/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-4-maverick. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama 4 Scout - Provider: OpenRouter - Page: https://llmeu.com/en/models/llama-4-scout/ - Context window: 1310720 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-4-scout. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Llama Guard 4 12B - Provider: OpenRouter - Page: https://llmeu.com/en/models/meta-llama-llama-guard-4-12b/ - Context window: 163840 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta-llama/llama-guard-4-12b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Glimmer 30B - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-glimmer-30b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-glimmer-30b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Spark 1.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-spark-1.1/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-spark-1.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Spark 1.2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-spark-1.2/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-spark-1.2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Spark 1.2 Contributor - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-spark-1.2-contributor/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-spark-1.2-contributor. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Spark 1.3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-spark-1.3/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-spark-1.3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Meta: Muse Spark 1.3 Contributor - Provider: OpenRouter - Page: https://llmeu.com/en/models/muse-spark-1.3-contributor/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as meta/muse-spark-1.3-contributor. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Microsoft: Phi 4 - Provider: OpenRouter - Page: https://llmeu.com/en/models/microsoft-phi-4/ - Context window: 16384 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as microsoft/phi-4. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-m1/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-minimax-m2/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M2-her - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-m2-her/ - Context window: 65536 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m2-her. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M2.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-m2.1/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m2.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M2.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-m2.5/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m2.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M2.7 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-m2.7/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m2.7. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax M3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-minimax-m3/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-m3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MiniMax: MiniMax-01 - Provider: OpenRouter - Page: https://llmeu.com/en/models/minimax-01/ - Context window: 1000192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as minimax/minimax-01. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Codestral 2508 - Provider: OpenRouter - Page: https://llmeu.com/en/models/codestral-2508/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/codestral-2508. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Codestral 2508 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/codestral-2508:batch/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/codestral-2508:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Devstral 2 2512 - Provider: OpenRouter - Page: https://llmeu.com/en/models/devstral-2512/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/devstral-2512. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Ministral 3 14B 2512 - Provider: OpenRouter - Page: https://llmeu.com/en/models/ministral-14b-2512/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/ministral-14b-2512. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Ministral 3 3B 2512 - Provider: OpenRouter - Page: https://llmeu.com/en/models/ministral-3b-2512/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/ministral-3b-2512. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Ministral 3 8B 2512 - Provider: OpenRouter - Page: https://llmeu.com/en/models/ministral-8b-2512/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/ministral-8b-2512. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Ministral 3 8B 2512 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/ministral-8b-2512:batch/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/ministral-8b-2512:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Large 3 2512 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-large-2512/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-large-2512. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Large 3 2512 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-large-2512:batch/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-large-2512:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Large 4 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-large-4-0/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-large-4-0. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Medium 3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-medium-3/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-medium-3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Medium 3.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-medium-3.1/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-medium-3.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Medium 3.1 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-medium-3.1:batch/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-medium-3.1:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Medium 3.5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-medium-3-5:batch/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-medium-3-5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Nemo - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-nemo/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-nemo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Small 3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-small-24b-instruct-2501/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-small-24b-instruct-2501. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Small 3.1 24B - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-small-3.1-24b-instruct/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-small-3.1-24b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Small 3.2 24B - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-small-3.2-24b-instruct/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-small-3.2-24b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Small 4 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-small-2603/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-small-2603. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Mistral Small 4 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-small-2603:batch/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-small-2603:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Saba - Provider: OpenRouter - Page: https://llmeu.com/en/models/mistral-saba/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/mistral-saba. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Mistral: Voxtral Small 24B 2507 - Provider: OpenRouter - Page: https://llmeu.com/en/models/voxtral-small-24b-2507/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as mistralai/voxtral-small-24b-2507. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2 0711 - Provider: OpenRouter - Page: https://llmeu.com/en/models/kimi-k2/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2 0905 - Provider: OpenRouter - Page: https://llmeu.com/en/models/kimi-k2-0905/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2-0905. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2 Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/moonshotai-kimi-k2-thinking/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/kimi-k2.5/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2.6 - Provider: OpenRouter - Page: https://llmeu.com/en/models/kimi-k2.6/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2.6. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MoonshotAI: Kimi K2.7 Code - Provider: OpenRouter - Page: https://llmeu.com/en/models/kimi-k2.7-code/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as moonshotai/kimi-k2.7-code. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Morph: Morph V3 Fast - Provider: OpenRouter - Page: https://llmeu.com/en/models/morph-v3-fast/ - Context window: 81920 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as morph/morph-v3-fast. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Morph: Morph V3 Large - Provider: OpenRouter - Page: https://llmeu.com/en/models/morph-v3-large/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as morph/morph-v3-large. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## MythoMax 13B - Provider: OpenRouter - Page: https://llmeu.com/en/models/mythomax-l2-13b/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as gryphe/mythomax-l2-13b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Nano 30B A3B - Provider: OpenRouter - Page: https://llmeu.com/en/models/nvidia-nemotron-3-nano-30b-a3b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-nano-30b-a3b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Nano Omni (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/nemotron-3-nano-omni-30b-a3b-reasoning:free/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Super - Provider: OpenRouter - Page: https://llmeu.com/en/models/nvidia-nemotron-3-super-120b-a12b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-super-120b-a12b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Super (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/nemotron-3-super-120b-a12b:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-super-120b-a12b:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Ultra - Provider: OpenRouter - Page: https://llmeu.com/en/models/nvidia-nemotron-3-ultra-550b-a55b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-ultra-550b-a55b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3 Ultra (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/nemotron-3-ultra-550b-a55b:free/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3-ultra-550b-a55b:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3.5 Lightning - Provider: OpenRouter - Page: https://llmeu.com/en/models/nemotron-3.5-lightning/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3.5-lightning. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## NVIDIA: Nemotron 3.5 Lightning (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/nemotron-3.5-lightning:free/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nvidia/nemotron-3.5-lightning:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Nex AGI: Nex-N2.5-Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/nex-n2.5-mini/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nex-agi/nex-n2.5-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Nex AGI: Nex-N2.5-Pro - Provider: OpenRouter - Page: https://llmeu.com/en/models/nex-n2.5-pro/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nex-agi/nex-n2.5-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Nous: Hermes 3 405B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/hermes-3-llama-3.1-405b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nousresearch/hermes-3-llama-3.1-405b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Nous: Hermes 3 70B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/hermes-3-llama-3.1-70b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nousresearch/hermes-3-llama-3.1-70b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Nous: Hermes 4 405B - Provider: OpenRouter - Page: https://llmeu.com/en/models/hermes-4-405b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as nousresearch/hermes-4-405b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT Audio Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-audio-mini/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-audio-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT Luna Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-luna-latest/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~openai/gpt-luna-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT Mini Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-mini-latest/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~openai/gpt-mini-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-3.5 Turbo - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-3.5-turbo/ - Context window: 16385 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-3.5-turbo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-3.5 Turbo (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-3.5-turbo:batch/ - Context window: 16385 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-3.5-turbo:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-3.5 Turbo (older v0613) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-3.5-turbo-0613/ - Context window: 4095 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-3.5-turbo-0613. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-3.5 Turbo Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-3.5-turbo-instruct/ - Context window: 4095 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-3.5-turbo-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4.1 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4.1:batch/ - Context window: 1047576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4.1:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4.1 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4.1-mini/ - Context window: 1047576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4.1-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4.1 Mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4.1-mini:batch/ - Context window: 1047576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4.1-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4.1 Nano - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4.1-nano/ - Context window: 1047576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4.1-nano. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4.1 Nano (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4.1-nano:batch/ - Context window: 1047576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4.1-nano:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4o (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4o:batch/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4o:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4o-mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4o-mini/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4o-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4o-mini (2024-07-18) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4o-mini-2024-07-18/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4o-mini-2024-07-18. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-4o-mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-4o-mini:batch/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-4o-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5-mini/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5 Mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5-mini:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5 Nano - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5-nano/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5-nano. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5 Nano (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5-nano:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5-nano:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.1 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.1:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.1:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.1-Codex-Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.1-codex-mini/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.1-codex-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.4 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.4-mini/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.4-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.4 Mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.4-mini:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.4-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.4 Nano - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.4-nano/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.4-nano. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.4 Nano (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.4-nano:batch/ - Context window: 400000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.4-nano:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Luna - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-luna/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-luna. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Luna (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-luna:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-luna:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Luna Pro - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-luna-pro/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-luna-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Luna Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-luna-pro:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-luna-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Sol (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-sol:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-sol:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-5.6 Sol Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-5.6-sol-pro:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-5.6-sol-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Luna - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-luna/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-luna. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Luna (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-luna:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-luna:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Luna Pro - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-luna-pro/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-luna-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Luna Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-luna-pro:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-luna-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Sol (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-sol:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-sol:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6 Sol Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6-sol-pro:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6-sol-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6.1 Sol (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6.1-sol:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6.1-sol:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: GPT-6.1 Sol Pro (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-6.1-sol-pro:batch/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-6.1-sol-pro:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: gpt-oss-120b - Provider: OpenRouter - Page: https://llmeu.com/en/models/openai-gpt-oss-120b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-oss-120b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: gpt-oss-120b (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-oss-120b:batch/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-oss-120b:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: gpt-oss-20b - Provider: OpenRouter - Page: https://llmeu.com/en/models/openai-gpt-oss-20b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-oss-20b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: gpt-oss-20b (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/gpt-oss-20b:batch/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-oss-20b:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: gpt-oss-safeguard-20b - Provider: OpenRouter - Page: https://llmeu.com/en/models/openai-gpt-oss-safeguard-20b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/gpt-oss-safeguard-20b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o3 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/o3:batch/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o3:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o3 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/o3-mini/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o3-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o3 Mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/o3-mini:batch/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o3-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o3 Mini High - Provider: OpenRouter - Page: https://llmeu.com/en/models/o3-mini-high/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o3-mini-high. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o4 Mini - Provider: OpenRouter - Page: https://llmeu.com/en/models/o4-mini/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o4-mini. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o4 Mini (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/o4-mini:batch/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o4-mini:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## OpenAI: o4 Mini High - Provider: OpenRouter - Page: https://llmeu.com/en/models/o4-mini-high/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as openai/o4-mini-high. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Pareto 26.10 Preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/pareto-26.10-preview/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as unbiased/pareto-26.10-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Perceptron: Perceptron Mk1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/perceptron-mk1/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as perceptron/perceptron-mk1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Perceptron: Perceptron Mk1.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/perceptron-mk1.5/ - Context window: 36864 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as perceptron/perceptron-mk1.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Perplexity: Sonar - Provider: OpenRouter - Page: https://llmeu.com/en/models/sonar/ - Context window: 127072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as perplexity/sonar. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Poolside: Laguna S 2.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/laguna-s-2.1/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as poolside/laguna-s-2.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Poolside: Laguna S 2.1 (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/laguna-s-2.1:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as poolside/laguna-s-2.1:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Poolside: Laguna XS 2.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/laguna-xs-2.1/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as poolside/laguna-xs-2.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Poolside: Laguna XS 2.1 (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/laguna-xs-2.1:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as poolside/laguna-xs-2.1:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## PrismML: Ternary Bonsai 2 27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/ternary-bonsai-2-27b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as prism-ml/ternary-bonsai-2-27b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen2.5 72B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-2.5-72b-instruct/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen-2.5-72b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen2.5 Coder 32B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-2.5-coder-32b-instruct/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen-2.5-coder-32b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen Plus 0728 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-plus-2025-07-28/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen-plus-2025-07-28. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen-Plus - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-plus/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen-plus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen2.5 7B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-2.5-7b-instruct/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen-2.5-7b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen2.5 VL 72B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen2.5-vl-72b-instruct/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen2.5-vl-72b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 14B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-14b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-14b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 235B A22B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-235b-a22b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-235b-a22b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 235B A22B Instruct 2507 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-235b-a22b-2507/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-235b-a22b-2507. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 235B A22B Thinking 2507 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-235b-a22b-thinking-2507/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-235b-a22b-thinking-2507. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 30B A3B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-30b-a3b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-30b-a3b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 30B A3B Instruct 2507 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-30b-a3b-instruct-2507/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-30b-a3b-instruct-2507. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 30B A3B Thinking 2507 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-30b-a3b-thinking-2507/ - Context window: 81920 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-30b-a3b-thinking-2507. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 32B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-32b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-32b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 8B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-8b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-8b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Coder 30B A3B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-coder-30b-a3b-instruct/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-coder-30b-a3b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Coder 480B A35B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-coder/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-coder. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Coder Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-coder-flash/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-coder-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Coder Next - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-coder-next/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-coder-next. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Coder Plus - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-coder-plus/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-coder-plus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Max - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-max/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-max. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Max Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-max-thinking/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-max-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Next 80B A3B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-next-80b-a3b-instruct/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-next-80b-a3b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 Next 80B A3B Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-next-80b-a3b-thinking/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-next-80b-a3b-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 235B A22B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-235b-a22b-instruct/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-235b-a22b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 235B A22B Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-235b-a22b-thinking/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-235b-a22b-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 30B A3B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-30b-a3b-instruct/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-30b-a3b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 30B A3B Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-30b-a3b-thinking/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-30b-a3b-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 32B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-32b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-32b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 8B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen-qwen3-vl-8b-instruct/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-8b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3 VL 8B Thinking - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3-vl-8b-thinking/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3-vl-8b-thinking. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5 397B A17B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-397b-a17b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-397b-a17b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5 Plus 2026-02-15 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-plus-02-15/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-plus-02-15. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5 Plus 2026-04-20 - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-plus-20260420/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-plus-20260420. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5-122B-A10B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-122b-a10b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-122b-a10b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5-27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-27b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-27b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5-35B-A3B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-35b-a3b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-35b-a3b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5-9B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-9b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-9b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.5-Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.5-flash-02-23/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.5-flash-02-23. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.6 27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.6-27b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.6-27b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.6 35B A3B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.6-35b-a3b/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.6-35b-a3b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.6 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.6-flash/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.6-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.6 Plus - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.6-plus/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.6-plus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.7 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.7-flash/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.7-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.7 Max - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.7-max/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.7-max. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.7 Plus - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.7-plus/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.7-plus. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.8 27B - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.8-27b/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.8-27b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.8 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.8-flash/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.8-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Qwen: Qwen3.8 Omni Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/qwen3.8-omni-flash/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as qwen/qwen3.8-omni-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## ReMM SLERP 13B - Provider: OpenRouter - Page: https://llmeu.com/en/models/remm-slerp-l2-13b/ - Context window: 6144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as undi95/remm-slerp-l2-13b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Reka Edge - Provider: OpenRouter - Page: https://llmeu.com/en/models/reka-edge/ - Context window: 16384 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as rekaai/reka-edge. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Reka Flash 3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/reka-flash-3/ - Context window: 65536 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as rekaai/reka-flash-3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Relace: Relace Apply 3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/relace-apply-3/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as relace/relace-apply-3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Relace: Relace Search - Provider: OpenRouter - Page: https://llmeu.com/en/models/relace-search/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as relace/relace-search. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Sakana: Sakana Namazu - Provider: OpenRouter - Page: https://llmeu.com/en/models/sakana-namazu/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as sakana/sakana-namazu. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Sao10K: Llama 3 8B Lunaris - Provider: OpenRouter - Page: https://llmeu.com/en/models/l3-lunaris-8b/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as sao10k/l3-lunaris-8b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Sao10K: Llama 3.1 Euryale 70B v2.2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/l3.1-euryale-70b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as sao10k/l3.1-euryale-70b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Sao10K: Llama 3.3 Euryale 70B - Provider: OpenRouter - Page: https://llmeu.com/en/models/l3.3-euryale-70b/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as sao10k/l3.3-euryale-70b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## SpaceXAI: Grok 4.20 - Provider: OpenRouter - Page: https://llmeu.com/en/models/grok-4.20/ - Context window: 2000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as x-ai/grok-4.20. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## SpaceXAI: Grok 4.20 Multi-Agent - Provider: OpenRouter - Page: https://llmeu.com/en/models/grok-4.20-multi-agent/ - Context window: 2000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as x-ai/grok-4.20-multi-agent. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## SpaceXAI: Grok 4.3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/grok-4.3/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as x-ai/grok-4.3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## SpaceXAI: Grok 4.3 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/grok-4.3:batch/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as x-ai/grok-4.3:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## SpaceXAI: Grok Build 0.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/grok-build-0.1/ - Context window: 256000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as x-ai/grok-build-0.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## StepFun: Step 3.5 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/step-3.5-flash/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as stepfun/step-3.5-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## StepFun: Step 3.7 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/step-3.7-flash/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as stepfun/step-3.7-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## StepFun: Step 5 Preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/step-5-preview/ - Context window: 1000000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as stepfun/step-5-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hunyuan A13B Instruct - Provider: OpenRouter - Page: https://llmeu.com/en/models/tencent-hunyuan-a13b-instruct/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hunyuan-a13b-instruct. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy-MT2-1.8B - Provider: OpenRouter - Page: https://llmeu.com/en/models/hy-mt2-1.8b/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy-mt2-1.8b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy-MT2-30B-A3B - Provider: OpenRouter - Page: https://llmeu.com/en/models/hy-mt2-30b-a3b/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy-mt2-30b-a3b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy-MT2-7B - Provider: OpenRouter - Page: https://llmeu.com/en/models/hy-mt2-7b/ - Context window: 8192 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy-mt2-7b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/tencent-hy3/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy3 preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/hy3-preview/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy3-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Tencent: Hy4 preview - Provider: OpenRouter - Page: https://llmeu.com/en/models/tencent-hy4-preview/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as tencent/hy4-preview. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## TheDrummer: Cydonia 24B V4.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/cydonia-24b-v4.1/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thedrummer/cydonia-24b-v4.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## TheDrummer: Skyfall 36B V2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/skyfall-36b-v2/ - Context window: 32768 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thedrummer/skyfall-36b-v2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## TheDrummer: UnslopNemo 12B - Provider: OpenRouter - Page: https://llmeu.com/en/models/unslopnemo-12b/ - Context window: 1024000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thedrummer/unslopnemo-12b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Thinking Machines: Inkling - Provider: OpenRouter - Page: https://llmeu.com/en/models/inkling/ - Context window: 524288 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thinkingmachines/inkling. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Thinking Machines: Inkling (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/inkling:free/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thinkingmachines/inkling:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Thinking Machines: Inkling Small - Provider: OpenRouter - Page: https://llmeu.com/en/models/inkling-small/ - Context window: 524288 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thinkingmachines/inkling-small. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Thinking Machines: Inkling Small (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/inkling-small:free/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as thinkingmachines/inkling-small:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Upstage: Solar Mini 4 - Provider: OpenRouter - Page: https://llmeu.com/en/models/solar-mini4/ - Context window: 524288 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as upstage/solar-mini4. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Upstage: Solar Pro 3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/solar-pro-3/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as upstage/solar-pro-3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Upstage: Solar Pro 4 - Provider: OpenRouter - Page: https://llmeu.com/en/models/solar-pro4/ - Context window: 524288 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as upstage/solar-pro4. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Venice: Uncensored - Provider: OpenRouter - Page: https://llmeu.com/en/models/dolphin-mistral-24b-venice-edition/ - Context window: 128000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as cognitivecomputations/dolphin-mistral-24b-venice-edition. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## WizardLM-2 8x22B - Provider: OpenRouter - Page: https://llmeu.com/en/models/wizardlm-2-8x22b/ - Context window: 65535 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as microsoft/wizardlm-2-8x22b. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Xiaomi: MiMo-V2.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/mimo-v2.5/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as xiaomi/mimo-v2.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Xiaomi: MiMo-V2.5-Pro - Provider: OpenRouter - Page: https://llmeu.com/en/models/mimo-v2.5-pro/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as xiaomi/mimo-v2.5-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Xiaomi: MiMo-V2.6-Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/mimo-v2.6-flash/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as xiaomi/mimo-v2.6-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Xiaomi: MiMo-V2.6-Pro - Provider: OpenRouter - Page: https://llmeu.com/en/models/mimo-v2.6-pro/ - Context window: 1050000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as xiaomi/mimo-v2.6-pro. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.5/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.5 Air - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.5-air/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.5-air. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.5V - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.5v/ - Context window: 65536 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.5v. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.6 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.6/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.6. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.6V - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.6v/ - Context window: 131072 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.6v. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.7 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.7/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.7. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 4.7 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-4.7-flash/ - Context window: 200000 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-4.7-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5 - Provider: OpenRouter - Page: https://llmeu.com/en/models/z-ai-glm-5/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5 Turbo - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5-turbo/ - Context window: 202752 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5-turbo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.1 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.1/ - Context window: 204800 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.1. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.2 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.2/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.2. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.3 - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.3/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.3. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.3 (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.3:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.3:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.3 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.3-flash/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.3-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.3 Flash (batch) - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.3-flash:batch/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.3-flash:batch. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5.3 FlashX - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5.3-flashx/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5.3-flashx. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM 5V Turbo - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-5v-turbo/ - Context window: 202752 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as z-ai/glm-5v-turbo. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM Flash Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-flash-latest/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~z-ai/glm-flash-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Z.ai: GLM Latest - Provider: OpenRouter - Page: https://llmeu.com/en/models/glm-latest/ - Context window: 1048576 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as ~z-ai/glm-latest. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## inclusionAI: Ling 3.0 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.0-flash/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inclusionai/ling-3.0-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## inclusionAI: Ling 3.0 Flash Fin - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.0-flash-fin/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inclusionai/ling-3.0-flash-fin. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## inclusionAI: Ling 3.0 Flash Sante - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.0-flash-sante/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inclusionai/ling-3.0-flash-sante. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## inclusionAI: Ling 3.0 Flash VL - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.0-flash-vl/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inclusionai/ling-3.0-flash-vl. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## inclusionAI: Ling 3.1 Flash - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.1-flash/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: hosted on LLM EU Served through OpenRouter as inclusionai/ling-3.1-flash. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## Aya 23 8B - Provider: Cohere - Page: https://llmeu.com/en/models/aya-23-8b/ - Context window: 8192 - Parameters: 8.03B - Licence: cc-by-nc-4.0 (https://cohere.com/c4ai-cc-by-nc-license) - Open weights: yes - Release date: 2024-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Aya 23 8B is the 8B member of Cohere Labs’ Aya 23 release, covering 23 languages under CC-BY-NC-4.0 with a context length of 8192 tokens. It is the multilingual baseline that appears in the OpenGPT-X European evaluation, where it scores 0.485 on EU21, below Llama 3.1 8B Instruct and Mistral 7B Instruct v0.3. It is superseded by Aya Expanse for most multilingual work. Sources: https://huggingface.co/CohereLabs/aya-23-8B, https://arxiv.org/abs/2405.15032 ## Aya Expanse 8B - Provider: Cohere - Page: https://llmeu.com/en/models/aya-expanse-8b/ - Context window: 8192 - Parameters: 8.03B - Licence: cc-by-nc-4.0 (https://cohere.com/c4ai-cc-by-nc-license) - Open weights: yes - Release date: 2024-10-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Aya Expanse 8B is Cohere Labs’ multilingual instruction model covering 23 languages, released under CC-BY-NC-4.0. Its metadata declares 23 languages, among them German, French, Spanish, Italian, Portuguese, Dutch, Polish, Czech, Romanian, Greek and Ukrainian, and the model card states a context length of 8K. Non-commercial use only: the weights may not be served commercially. Sources: https://huggingface.co/CohereLabs/aya-expanse-8b, https://cohere.com/c4ai-cc-by-nc-license ## Blip2 Flan T5 Xl - Provider: Salesforce - Page: https://llmeu.com/en/models/blip2-flan-t5-xl/ - Context window: 0 - Parameters: 3.9B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2023-02-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Blip2 Flan T5 Xl is an open-weight vision-language model from Salesforce with 3.9B parameters, published under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Salesforce/blip2-flan-t5-xl, https://huggingface.co/api/models/Salesforce/blip2-flan-t5-xl?blobs=true, https://opensource.org/license/mit ## Blip2 Opt 2.7B - Provider: Salesforce - Page: https://llmeu.com/en/models/blip2-opt-2-7b/ - Context window: 0 - Parameters: 3.7B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2023-02-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Blip2 Opt 2.7B is an open-weight vision-language model from Salesforce with 3.7B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Salesforce/blip2-opt-2.7b, https://huggingface.co/api/models/Salesforce/blip2-opt-2.7b?blobs=true, https://opensource.org/license/mit ## Bloom - Provider: BigScience (BLOOM) - Page: https://llmeu.com/en/models/bloom/ - Context window: 0 - Parameters: 176.2B - Licence: bigscience-bloom-rail-1.0 (https://bigscience.huggingface.co/blog/the-bigscience-rail-license) - Open weights: yes - Release date: 2022-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Bloom is a very large open-weight language model from BigScience (BLOOM): 176.2B parameters under bigscience-bloom-rail-1.0. The published configuration records no context window, so this card carries none. The repository was created in May 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 45 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/bigscience/bloom, https://huggingface.co/api/models/bigscience/bloom?blobs=true, https://bigscience.huggingface.co/blog/the-bigscience-rail-license ## Bloom 1b1 - Provider: BigScience (BLOOM) - Page: https://llmeu.com/en/models/bloom-1b1/ - Context window: 0 - Parameters: 1.1B - Licence: bigscience-bloom-rail-1.0 (https://bigscience.huggingface.co/blog/the-bigscience-rail-license) - Open weights: yes - Release date: 2022-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Bloom 1b1 is a small open-weight language model from BigScience (BLOOM): 1.1B parameters under bigscience-bloom-rail-1.0. The repository was created in May 2022 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 48 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/bigscience/bloom-1b1, https://huggingface.co/api/models/bigscience/bloom-1b1?blobs=true, https://bigscience.huggingface.co/blog/the-bigscience-rail-license ## Bloom 3B - Provider: BigScience (BLOOM) - Page: https://llmeu.com/en/models/bloom-3b/ - Context window: 0 - Parameters: 3B - Licence: bigscience-bloom-rail-1.0 (https://bigscience.huggingface.co/blog/the-bigscience-rail-license) - Open weights: yes - Release date: 2022-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Bloom 3B is a mid-size open-weight language model from BigScience (BLOOM): 3B parameters under bigscience-bloom-rail-1.0. The published configuration records no context window, so this card carries none. The repository was created in May 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 47 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/bigscience/bloom-3b, https://huggingface.co/api/models/bigscience/bloom-3b?blobs=true, https://bigscience.huggingface.co/blog/the-bigscience-rail-license ## Bloom 560M - Provider: BigScience (BLOOM) - Page: https://llmeu.com/en/models/bloom-560m/ - Context window: 0 - Parameters: 0.6B - Licence: bigscience-bloom-rail-1.0 (https://bigscience.huggingface.co/blog/the-bigscience-rail-license) - Open weights: yes - Release date: 2022-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Bloom 560M is a small open-weight language model from BigScience (BLOOM): 0.6B parameters under bigscience-bloom-rail-1.0. The repository was created in May 2022 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 48 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/bigscience/bloom-560m, https://huggingface.co/api/models/bigscience/bloom-560m?blobs=true, https://bigscience.huggingface.co/blog/the-bigscience-rail-license ## Bloomz 560M - Provider: BigScience (BLOOM) - Page: https://llmeu.com/en/models/bloomz-560m/ - Context window: 0 - Parameters: 0.6B - Licence: bigscience-bloom-rail-1.0 (https://bigscience.huggingface.co/blog/the-bigscience-rail-license) - Open weights: yes - Release date: 2022-10-08 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Bloomz 560M is a small open-weight language model from BigScience (BLOOM): 0.6B parameters under bigscience-bloom-rail-1.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in October 2022 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 45 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/bigscience/bloomz-560m, https://huggingface.co/api/models/bigscience/bloomz-560m?blobs=true, https://bigscience.huggingface.co/blog/the-bigscience-rail-license ## Codestral 25.08 - Provider: Mistral AI - Page: https://llmeu.com/en/models/codestral-25-08/ - Context window: 131072 - Parameters: unverified - Licence: proprietary (https://legal.mistral.ai/ai-governance/models/codestral) - Open weights: no - Release date: 2025-07-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Codestral is Mistral’s code-completion model, served through the Mistral API rather than published as weights. It supports fill-in-the-middle, tool calling and document question answering over a 128k-token context window. The parameter count is not published, and no price is quoted here: this deployment serves nothing, so a rate nobody can be charged is not a fact this page should carry. Sources: https://docs.mistral.ai/models/codestral-25-08, https://mistral.ai/news/codestral-25-08 ## Command A+ (05-2026) - Provider: Cohere - Page: https://llmeu.com/en/models/command-a-plus-05-2026/ - Context window: 131072 - Parameters: 218.75B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-05-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Command A+ (command-a-plus-05-2026) is Cohere’s open-weights 219B-parameter model under Apache-2.0, taking text and images and declaring 48 languages. Cohere publishes FP8 and W4A4 checkpoints with the hardware each needs; at BF16 the weights need a multi-GPU node. Sources: https://huggingface.co/CohereLabs/command-a-plus-05-2026-bf16 ## Command R (08-2024) - Provider: Cohere - Page: https://llmeu.com/en/models/command-r-08-2024/ - Context window: 131072 - Parameters: 32.3B - Licence: cc-by-nc-4.0 (https://cohere.com/c4ai-cc-by-nc-license) - Open weights: yes - Release date: 2024-08-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Command R (08-2024) is a 32B retrieval-augmented-generation model with a 128K-token context window, published under CC-BY-NC-4.0. That licence allows research and non-commercial use of the weights but excludes commercial serving and requires accepting Cohere’s acceptable-use policy. The model is built for citation-backed answers over retrieved documents and for tool use. Sources: https://huggingface.co/CohereLabs/c4ai-command-r-08-2024, https://cohere.com/c4ai-cc-by-nc-license ## DeepSeek Coder V2 Lite Instruct - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-coder-v2-lite-instruct/ - Context window: 0 - Parameters: 15.7B - Licence: other (https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-06-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek Coder V2 Lite Instruct is an open-weight language model from DeepSeek with 15.7B parameters, published under deepseek-license (custom, tagged "other"). The repository was created in June 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct, https://huggingface.co/api/models/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct?blobs=true, https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct/blob/main/LICENSE ## DeepSeek OCR - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-ocr/ - Context window: 0 - Parameters: 3.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-10-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek OCR is an open-weight vision-language model from DeepSeek with 3.3B parameters, published under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/deepseek-ai/DeepSeek-OCR, https://huggingface.co/api/models/deepseek-ai/DeepSeek-OCR?blobs=true, https://opensource.org/license/mit ## DeepSeek OCR 2 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-ocr-2/ - Context window: 0 - Parameters: 3.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek OCR 2 is an open-weight vision-language model from DeepSeek with 3.4B parameters, published under Apache-2.0. The repository was created in January 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/deepseek-ai/DeepSeek-OCR-2, https://huggingface.co/api/models/deepseek-ai/DeepSeek-OCR-2?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## DeepSeek R1 0528 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-0528/ - Context window: 0 - Parameters: 684.5B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-05-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek R1 0528 is an open-weight language model from DeepSeek with 684.5B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-0528, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-0528?blobs=true, https://opensource.org/license/mit ## DeepSeek R1 0528 Qwen3 8B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-0528-qwen3-8b/ - Context window: 0 - Parameters: 8.2B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-05-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. DeepSeek R1 0528 Qwen3 8B is a mid-size open-weight language model from DeepSeek: 8.2B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B?blobs=true, https://opensource.org/license/mit ## DeepSeek R1 Distill Llama 8B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-distill-llama-8b/ - Context window: 0 - Parameters: 8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. DeepSeek R1 Distill Llama 8B is a mid-size open-weight language model from DeepSeek: 8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in January 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-Distill-Llama-8B?blobs=true, https://opensource.org/license/mit ## DeepSeek R1 Distill Qwen 1.5B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-distill-qwen-1-5b/ - Context window: 0 - Parameters: 1.8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek R1 Distill Qwen 1.5B is an open-weight language model from DeepSeek with 1.8B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B?blobs=true, https://opensource.org/license/mit ## DeepSeek R1 Distill Qwen 14B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-distill-qwen-14b/ - Context window: 0 - Parameters: 14.8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek R1 Distill Qwen 14B is an open-weight language model from DeepSeek with 14.8B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B?blobs=true, https://opensource.org/license/mit ## DeepSeek R1 Distill Qwen 7B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-distill-qwen-7b/ - Context window: 0 - Parameters: 7.6B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. DeepSeek R1 Distill Qwen 7B is an open-weight language model from DeepSeek with 7.6B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in January 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B, https://huggingface.co/api/models/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B?blobs=true, https://opensource.org/license/mit ## DeepSeek V2 Lite - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v2-lite/ - Context window: 0 - Parameters: 15.7B - Licence: other (https://github.com/deepseek-ai/DeepSeek-V2/blob/main/LICENSE-MODEL) - Open weights: yes - Release date: 2024-05-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek V2 Lite is a large open-weight language model from DeepSeek: 15.7B parameters under deepseek (custom, tagged "other"). The repository was created in May 2024 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V2-Lite?blobs=true, https://github.com/deepseek-ai/DeepSeek-V2/blob/main/LICENSE-MODEL ## DeepSeek V3 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v3/ - Context window: 0 - Parameters: unverified - Licence: unknown (https://huggingface.co/deepseek-ai/DeepSeek-V3/blob/main/LICENSE) - Open weights: yes - Release date: 2024-12-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek V3 is an open-weight language model from DeepSeek with an unpublished parameter count, published under unknown. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V3, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V3?blobs=true, https://huggingface.co/deepseek-ai/DeepSeek-V3/blob/main/LICENSE ## DeepSeek V3 0324 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v3-0324/ - Context window: 0 - Parameters: 684.5B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-03-24 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. DeepSeek V3 0324 is an open-weight language model from DeepSeek with 684.5B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in March 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V3-0324, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V3-0324?blobs=true, https://opensource.org/license/mit ## DeepSeek V3.1 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v3-1/ - Context window: 0 - Parameters: 684.5B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-08-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek V3.1 is an open-weight language model from DeepSeek with 684.5B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V3.1, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V3.1?blobs=true, https://opensource.org/license/mit ## DeepSeek V4 Flash - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v4-flash/ - Context window: 0 - Parameters: 290.9B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-04-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek V4 Flash is published by DeepSeek as open weights with 290.9B parameters under MIT. The repository was created in April 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4-Flash?blobs=true, https://opensource.org/license/mit ## DeepSeek V4 Pro - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v4-pro/ - Context window: 0 - Parameters: 1598.8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-04-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. DeepSeek V4 Pro is published by DeepSeek as open weights with 1598.8B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4-Pro?blobs=true, https://opensource.org/license/mit ## DeepSeek V4.1 Flash - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v4-1-flash/ - Context window: 0 - Parameters: 763.2B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-09-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek V4.1 Flash is published by DeepSeek as open weights with 763.2B parameters under MIT. The repository was created in September 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash, https://huggingface.co/api/models/deepseek-ai/DeepSeek-V4.1-Flash?blobs=true, https://opensource.org/license/mit ## DeepSeek-R1 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1/ - Context window: 131072 - Parameters: 684.49B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek-R1 is the reasoning model that made open-weight chain-of-thought training widely reproducible; it has 671B total and 37B activated parameters and is published under the MIT licence. The model card documents a context length of 128K and reports 90.8 on MMLU, 84.0 on MMLU-Pro and 79.8 on AIME 2024 for R1. It is a text-only model and has been superseded for most tasks by DeepSeek-V3.2 and V4. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1, https://arxiv.org/abs/2501.12948 ## DeepSeek-R1-Distill-Qwen-32B - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-r1-distill-qwen-32b/ - Context window: 131072 - Parameters: 32.76B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-01-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek-R1-Distill-Qwen-32B is a 32B dense model distilled from DeepSeek-R1 into a Qwen2.5 base, published under the MIT licence. It keeps the reasoning behaviour of R1 at a size that fits on one 80GB accelerator, with a configuration maximum of 131,072 tokens. DeepSeek documents it as a distillation, so its ceiling is the teacher model’s reasoning, not a new training run. Sources: https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-32B ## DeepSeek-V3.2 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v3-2/ - Context window: 163840 - Parameters: 685.36B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-12-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek-V3.2 is a 685B mixture-of-experts model released under the MIT licence, with a configuration maximum of 163,840 tokens. Its main architectural change is DeepSeek Sparse Attention, which the vendor describes as reducing the computational cost of long-context inference. It is a text-only model and needs a multi-node deployment at full precision. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V3.2 ## DeepSeek-V4-Flash (0731) - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-v4-flash-0731/ - Context window: 1048576 - Parameters: 304.18B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-07-31 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report DeepSeek-V4-Flash is a mixture-of-experts model with 304B total parameters, published under the MIT licence with a context window of 1,048,576 tokens in its configuration. It exposes three reasoning-effort levels and the vendor recommends allowing up to 384K output tokens at the high and max settings. DeepSeek states it outperforms its own DeepSeek-V4-Pro preview on the benchmarks listed in the model card, which are self-reported. Sources: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731 ## Deepseek Coder 7B Instruct V1.5 - Provider: DeepSeek - Page: https://llmeu.com/en/models/deepseek-coder-7b-instruct-v1-5/ - Context window: 0 - Parameters: 6.9B - Licence: other (https://huggingface.co/deepseek-ai/deepseek-coder-7b-instruct-v1.5/blob/main/LICENSE) - Open weights: yes - Release date: 2024-01-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Deepseek Coder 7B Instruct V1.5 is an open-weight language model from DeepSeek with 6.9B parameters, published under deepseek (custom, tagged "other"). The repository was created in January 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/deepseek-ai/deepseek-coder-7b-instruct-v1.5, https://huggingface.co/api/models/deepseek-ai/deepseek-coder-7b-instruct-v1.5?blobs=true, https://huggingface.co/deepseek-ai/deepseek-coder-7b-instruct-v1.5/blob/main/LICENSE ## Devstral Small 2 (24B) - Provider: Mistral AI - Page: https://llmeu.com/en/models/devstral-small-2-24b/ - Context window: 262144 - Parameters: 24.01B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Devstral Small 2 24B (devstral-small-2-24b) is Mistral’s Apache-2.0 coding model tuned with All Hands AI for agents that edit files and call tools. Text-only, a 256k-token context window, and a model card reporting 68.0% on SWE-bench Verified and 55.7% on SWE-bench Multilingual. Sources: https://huggingface.co/mistralai/Devstral-Small-2-24B-Instruct-2512 ## Diffusiongemma 26B A4B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/diffusiongemma-26b-a4b-it/ - Context window: 0 - Parameters: 25.8B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-06-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Diffusiongemma 26B A4B It is an open-weight vision-language model from Google DeepMind with 25.8B parameters, published under Apache-2.0. The repository was created in June 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/diffusiongemma-26B-A4B-it, https://huggingface.co/api/models/google/diffusiongemma-26B-A4B-it?blobs=true, https://ai.google.dev/gemma/docs/gemma_4_license ## ERNIE 4.5 0.3B - Provider: Baidu - Page: https://llmeu.com/en/models/ernie-4-5-0-3b/ - Context window: 0 - Parameters: 0.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report ERNIE 4.5 0.3B is an open-weight language model from Baidu with 0.4B parameters, published under Apache-2.0. The repository was created in June 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/baidu/ERNIE-4.5-0.3B-PT, https://huggingface.co/api/models/baidu/ERNIE-4.5-0.3B-PT?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## ERNIE 4.5 21B-A3B Thinking - Provider: Baidu - Page: https://llmeu.com/en/models/ernie-4-5-21b-a3b-thinking/ - Context window: 0 - Parameters: 21.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-08 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report ERNIE 4.5 21B-A3B Thinking is an open-weight language model from Baidu with 21.8B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/baidu/ERNIE-4.5-21B-A3B-Thinking, https://huggingface.co/api/models/baidu/ERNIE-4.5-21B-A3B-Thinking?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## ERNIE 4.5 VL 28B-A3B - Provider: Baidu - Page: https://llmeu.com/en/models/ernie-4-5-vl-28b-a3b/ - Context window: 0 - Parameters: 29.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report ERNIE 4.5 VL 28B-A3B is an open-weight vision-language model from Baidu with 29.4B parameters, published under Apache-2.0. The repository was created in June 2025 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/baidu/ERNIE-4.5-VL-28B-A3B-PT, https://huggingface.co/api/models/baidu/ERNIE-4.5-VL-28B-A3B-PT?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## EmbeddingGemma 300M - Provider: Google DeepMind - Page: https://llmeu.com/en/models/embeddinggemma-300m/ - Context window: 2048 - Parameters: 0.3B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2025-07-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report EmbeddingGemma is a 308M-parameter text embedding model with a 768-dimensional output that can be truncated to 512, 256 or 128 dimensions through Matryoshka representation learning. The model card states a maximum input length of 2048 tokens and training data in over 100 spoken languages. It is released under Google’s Gemma terms rather than an OSI licence. Sources: https://huggingface.co/google/embeddinggemma-300m, https://arxiv.org/abs/2509.20354 ## EuroLLM-1.7B-Instruct - Provider: EuroLLM (utter-project) - Page: https://llmeu.com/en/models/eurollm-1-7b-instruct/ - Context window: 4096 - Parameters: 1.66B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-08-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report EuroLLM 1.7B Instruct is the smallest EuroLLM model, at 1.66B parameters under Apache-2.0. The card records four trillion training tokens across the same 35 languages as the larger sizes and a 4096-token context window, and the footprint runs on CPU-class or very small GPU hardware for translation and lighter multilingual tasks. Sources: https://huggingface.co/utter-project/EuroLLM-1.7B-Instruct ## EuroLLM-22B-Instruct (2512) - Provider: EuroLLM (utter-project) - Page: https://llmeu.com/en/models/eurollm-22b-instruct-2512/ - Context window: 32768 - Parameters: 22.64B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-12-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report EuroLLM 22B (eurollm-22b-instruct-2512) is the largest released EuroLLM checkpoint, at 22.6B parameters under Apache-2.0. It has a 32k-token context window, covers the same European language set as the 9B version, and is strongest in English, German, French, Spanish, Italian and Portuguese. Sources: https://huggingface.co/utter-project/EuroLLM-22B-Instruct-2512 ## EuroLLM-9B-Instruct (2512) - Provider: EuroLLM (utter-project) - Page: https://llmeu.com/en/models/eurollm-9b-instruct-2512/ - Context window: 32768 - Parameters: 9.15B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report EuroLLM 9B (eurollm-9b-instruct-2512) is a 9.15B fine-tune for European languages. It covers the 24 official EU languages plus Catalan, Galician, Turkish, Arabic, Chinese, Japanese, Korean, Russian and Hindi, under Apache-2.0 with a 32k-token context window and a post-training recipe taken from the 22B release. Sources: https://huggingface.co/utter-project/EuroLLM-9B-Instruct-2512 ## Falcon 7B - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-7b/ - Context window: 0 - Parameters: 7.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-04-24 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Falcon 7B is published by Technology Innovation Institute (TII) as open weights with 7.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/falcon-7b, https://huggingface.co/api/models/tiiuae/falcon-7b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Falcon 7B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-7b-instruct/ - Context window: 0 - Parameters: 7.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-04-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Falcon 7B Instruct is published by Technology Innovation Institute (TII) as open weights with 7.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/falcon-7b-instruct, https://huggingface.co/api/models/tiiuae/falcon-7b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Falcon H1 0.5B Base - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-h1-0-5b-base/ - Context window: 0 - Parameters: 0.5B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2025-05-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Falcon H1 0.5B Base is a small open-weight language model from Technology Innovation Institute (TII): 0.5B parameters under falcon-llm-license (custom, tagged "other"). The repository was created in May 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/Falcon-H1-0.5B-Base, https://huggingface.co/api/models/tiiuae/Falcon-H1-0.5B-Base?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Falcon H1 1.5B Deep Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-h1-1-5b-deep-instruct/ - Context window: 0 - Parameters: 1.6B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2025-05-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Falcon H1 1.5B Deep Instruct is a small open-weight language model from Technology Innovation Institute (TII): 1.6B parameters under falcon-llm-license (custom, tagged "other"). The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tiiuae/Falcon-H1-1.5B-Deep-Instruct, https://huggingface.co/api/models/tiiuae/Falcon-H1-1.5B-Deep-Instruct?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Falcon H1 1.5B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-h1-1-5b-instruct/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/tiiuae/Falcon-H1-1.5B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2025-05-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Falcon H1 1.5B Instruct is a unstated size open-weight language model from Technology Innovation Institute (TII): an unpublished parameter count under other. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 18 languages, 11 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/Falcon-H1-1.5B-Instruct, https://huggingface.co/api/models/tiiuae/Falcon-H1-1.5B-Instruct?blobs=true, https://huggingface.co/tiiuae/Falcon-H1-1.5B-Instruct/blob/main/LICENSE ## Falcon H1 3B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-h1-3b-instruct/ - Context window: 0 - Parameters: 3.1B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2025-05-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Falcon H1 3B Instruct is a mid-size open-weight language model from Technology Innovation Institute (TII): 3.1B parameters under falcon-llm-license (custom, tagged "other"). The repository was created in May 2025 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 18 languages, 11 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/Falcon-H1-3B-Instruct, https://huggingface.co/api/models/tiiuae/Falcon-H1-3B-Instruct?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Falcon H1 7B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-h1-7b-instruct/ - Context window: 0 - Parameters: 7.6B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2025-05-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Falcon H1 7B Instruct is a mid-size open-weight language model from Technology Innovation Institute (TII): 7.6B parameters under falcon-llm-license (custom, tagged "other"). The repository was created in May 2025 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 18 languages, 11 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/Falcon-H1-7B-Instruct, https://huggingface.co/api/models/tiiuae/Falcon-H1-7B-Instruct?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Falcon Mamba 7B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon-mamba-7b-instruct/ - Context window: 0 - Parameters: 7.3B - Licence: other (https://falconllm.tii.ae/falcon-mamba-7b-terms-and-conditions.html) - Open weights: yes - Release date: 2024-07-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Falcon Mamba 7B Instruct is published by Technology Innovation Institute (TII) as open weights with 7.3B parameters under falcon-mamba-7b-license (custom, tagged "other"). The published configuration records no context window, so this card carries none. The repository was created in July 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/falcon-mamba-7b-instruct, https://huggingface.co/api/models/tiiuae/falcon-mamba-7b-instruct?blobs=true, https://falconllm.tii.ae/falcon-mamba-7b-terms-and-conditions.html ## Falcon3 10B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon3-10b-instruct/ - Context window: 0 - Parameters: 10.3B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2024-12-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Falcon3 10B Instruct is published by Technology Innovation Institute (TII) as open weights with 10.3B parameters under falcon-llm-license (custom, tagged "other"). The repository was created in December 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tiiuae/Falcon3-10B-Instruct, https://huggingface.co/api/models/tiiuae/Falcon3-10B-Instruct?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Falcon3 7B Instruct - Provider: Technology Innovation Institute (TII) - Page: https://llmeu.com/en/models/falcon3-7b-instruct/ - Context window: 0 - Parameters: 7.5B - Licence: other (https://falconllm.tii.ae/falcon-terms-and-conditions.html) - Open weights: yes - Release date: 2024-11-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Falcon3 7B Instruct is published by Technology Innovation Institute (TII) as open weights with 7.5B parameters under falcon-llm-license (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in November 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 4 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tiiuae/Falcon3-7B-Instruct, https://huggingface.co/api/models/tiiuae/Falcon3-7B-Instruct?blobs=true, https://falconllm.tii.ae/falcon-terms-and-conditions.html ## Flan T5 Base - Provider: Google DeepMind - Page: https://llmeu.com/en/models/flan-t5-base/ - Context window: 0 - Parameters: 0.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-10-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Flan T5 Base is published by Google DeepMind as open weights with 0.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 6 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/google/flan-t5-base, https://huggingface.co/api/models/google/flan-t5-base?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Flan T5 Small - Provider: Google DeepMind - Page: https://llmeu.com/en/models/flan-t5-small/ - Context window: 0 - Parameters: 0.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-10-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Flan T5 Small is published by Google DeepMind as open weights with 0.1B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 6 languages, 4 of them among the languages the European Union publishes in. Sources: https://huggingface.co/google/flan-t5-small, https://huggingface.co/api/models/google/flan-t5-small?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Florence 2 Base Ft - Provider: Microsoft - Page: https://llmeu.com/en/models/florence-2-base-ft/ - Context window: 0 - Parameters: 0.2B - Licence: mit (https://huggingface.co/microsoft/Florence-2-base-ft/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-06-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Florence 2 Base Ft is an open-weight vision-language model from Microsoft with 0.2B parameters, published under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/microsoft/Florence-2-base-ft, https://huggingface.co/api/models/microsoft/Florence-2-base-ft?blobs=true, https://huggingface.co/microsoft/Florence-2-base-ft/resolve/main/LICENSE ## Florence 2 Large Ft - Provider: Microsoft - Page: https://llmeu.com/en/models/florence-2-large-ft/ - Context window: 0 - Parameters: 0.8B - Licence: mit (https://huggingface.co/microsoft/Florence-2-large-ft/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-06-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Florence 2 Large Ft is an open-weight vision-language model from Microsoft with 0.8B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in June 2024 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/microsoft/Florence-2-large-ft, https://huggingface.co/api/models/microsoft/Florence-2-large-ft?blobs=true, https://huggingface.co/microsoft/Florence-2-large-ft/resolve/main/LICENSE ## GLM 4.1V 9B Thinking - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-1v-9b-thinking/ - Context window: 0 - Parameters: 10.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-06-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. GLM 4.1V 9B Thinking is published by Z.ai (Zhipu AI) as open weights with 10.3B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.1V-9B-Thinking, https://huggingface.co/api/models/zai-org/GLM-4.1V-9B-Thinking?blobs=true, https://opensource.org/license/mit ## GLM 4.5 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-5/ - Context window: 0 - Parameters: 358.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-07-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM 4.5 is an open-weight language model from Z.ai (Zhipu AI) with 358.3B parameters, published under MIT. The repository was created in July 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.5, https://huggingface.co/api/models/zai-org/GLM-4.5?blobs=true, https://opensource.org/license/mit ## GLM 4.5 Air - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-5-air/ - Context window: 0 - Parameters: 110.5B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-07-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. GLM 4.5 Air is an open-weight language model from Z.ai (Zhipu AI) with 110.5B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in July 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.5-Air, https://huggingface.co/api/models/zai-org/GLM-4.5-Air?blobs=true, https://opensource.org/license/mit ## GLM 4.5V - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-5v/ - Context window: 0 - Parameters: 107.7B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-08-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. GLM 4.5V is a very large open-weight vision-language model from Z.ai (Zhipu AI): 107.7B parameters under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in August 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.5V, https://huggingface.co/api/models/zai-org/GLM-4.5V?blobs=true, https://opensource.org/license/mit ## GLM 4.6 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-6/ - Context window: 0 - Parameters: 356.8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-09-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. GLM 4.6 is published by Z.ai (Zhipu AI) as open weights with 356.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in September 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.6, https://huggingface.co/api/models/zai-org/GLM-4.6?blobs=true, https://opensource.org/license/mit ## GLM 4.6V Flash - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-6v-flash/ - Context window: 0 - Parameters: 10.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-12-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. GLM 4.6V Flash is an open-weight vision-language model from Z.ai (Zhipu AI) with 10.3B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in December 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-4.6V-Flash, https://huggingface.co/api/models/zai-org/GLM-4.6V-Flash?blobs=true, https://opensource.org/license/mit ## GLM 5 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-5/ - Context window: 0 - Parameters: 753.9B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-02-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM 5 is published by Z.ai (Zhipu AI) as open weights with 753.9B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-5, https://huggingface.co/api/models/zai-org/GLM-5?blobs=true, https://opensource.org/license/mit ## GLM 5.1 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-5-1/ - Context window: 0 - Parameters: 753.9B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-04-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM 5.1 is published by Z.ai (Zhipu AI) as open weights with 753.9B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-5.1, https://huggingface.co/api/models/zai-org/GLM-5.1?blobs=true, https://opensource.org/license/mit ## GLM 5.2 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-5-2/ - Context window: 0 - Parameters: 753.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-06-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM 5.2 is published by Z.ai (Zhipu AI) as open weights with 753.3B parameters under MIT. The repository was created in June 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-5.2, https://huggingface.co/api/models/zai-org/GLM-5.2?blobs=true, https://opensource.org/license/mit ## GLM 5.3 Flash - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-5-3-flash/ - Context window: 0 - Parameters: 321.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-08-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. GLM 5.3 Flash is a very large open-weight vision-language model from Z.ai (Zhipu AI): 321.3B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in August 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-5.3-Flash, https://huggingface.co/api/models/zai-org/GLM-5.3-Flash?blobs=true, https://opensource.org/license/mit ## GLM Z1 32B 0414 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-z1-32b-0414/ - Context window: 0 - Parameters: 32.6B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-04-08 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. GLM Z1 32B 0414 is an open-weight language model from Z.ai (Zhipu AI) with 32.6B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/GLM-Z1-32B-0414, https://huggingface.co/api/models/zai-org/GLM-Z1-32B-0414?blobs=true, https://opensource.org/license/mit ## GLM-4 9B Chat 1M - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-9b-chat-1m/ - Context window: 0 - Parameters: 9.5B - Licence: other (https://huggingface.co/THUDM/glm-4-9b-chat-1m/blob/main/LICENSE) - Open weights: yes - Release date: 2024-06-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM-4 9B Chat 1M is an open-weight language model from Z.ai (Zhipu AI) with 9.5B parameters, published under glm-4 (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/glm-4-9b-chat-1m, https://huggingface.co/api/models/zai-org/glm-4-9b-chat-1m?blobs=true, https://huggingface.co/THUDM/glm-4-9b-chat-1m/blob/main/LICENSE ## GLM-4.7-Flash - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-7-flash/ - Context window: 202752 - Parameters: 31.22B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-01-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM-4.7-Flash is a 30B-A3B mixture-of-experts model with 31.2B parameters, released under the MIT licence with a configuration maximum of 202,752 tokens. The model card reports 91.6 on AIME 25 and 59.2 on SWE-bench Verified, with a comparison table against Qwen3-30B-A3B-Thinking and gpt-oss-20b. It is positioned as a lightweight deployment option in the 30B class. Sources: https://huggingface.co/zai-org/GLM-4.7-Flash, https://z.ai/blog/glm-4.7 ## GLM-5.3 - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-5-3/ - Context window: 1048576 - Parameters: 753.33B - Licence: other (https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE) - Open weights: yes - Release date: 2026-08-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM-5.3 is Z.ai’s largest open-weights model at 753B parameters, with a configuration maximum of 1,048,576 tokens and a bespoke "glm-5.3" licence rather than MIT. The vendor states it shares its base model with GLM-5.2 and that all gains come from post-training, and reports 88.8% on Terminal-Bench 2.1 and 64.5% on HLE with tools. The weights are large enough to need a multi-node deployment. Sources: https://huggingface.co/zai-org/GLM-5.3, https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE ## GLM-OCR - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-ocr/ - Context window: 131072 - Parameters: 1.33B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-01-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report GLM-OCR is an OCR model for extracting text from document images: a CogViT vision encoder with a GLM-0.5B decoder. Its MIT licence allows commercial self-hosting, its configuration allows 131,072 text positions for long documents, and it runs under vLLM, SGLang or Ollama — and the card records that the published safetensors total 1.33B parameters against the 0.9B the model card states. Sources: https://huggingface.co/zai-org/GLM-OCR ## Gemma 2 2B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-2-2b-it/ - Context window: 0 - Parameters: 2.6B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2024-07-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 2 2B It is a mid-size open-weight language model from Google DeepMind: 2.6B parameters under gemma. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-2-2b-it, https://huggingface.co/api/models/google/gemma-2-2b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 2 9B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-2-9b-it/ - Context window: 0 - Parameters: 9.2B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2024-06-24 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Gemma 2 9B It is a mid-size open-weight language model from Google DeepMind: 9.2B parameters under gemma. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in June 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-2-9b-it, https://huggingface.co/api/models/google/gemma-2-9b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 3 12B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-3-12b-it/ - Context window: 0 - Parameters: 12.2B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2025-03-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 3 12B It is an open-weight vision-language model from Google DeepMind with 12.2B parameters, published under gemma. The repository was created in March 2025 and is mirrored on Hugging Face. It accepts text and images and answers in text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-3-12b-it, https://huggingface.co/api/models/google/gemma-3-12b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 3 1B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-3-1b-it/ - Context window: 0 - Parameters: 1B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2025-03-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 3 1B It is an open-weight language model from Google DeepMind with 1B parameters, published under gemma. The repository was created in March 2025 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-3-1b-it, https://huggingface.co/api/models/google/gemma-3-1b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 3 27B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-3-27b-it/ - Context window: 0 - Parameters: 27.4B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2025-03-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 3 27B It is an open-weight vision-language model from Google DeepMind with 27.4B parameters, published under gemma. The repository was created in March 2025 and is mirrored on Hugging Face. It accepts text and images and answers in text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-3-27b-it, https://huggingface.co/api/models/google/gemma-3-27b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 3 4B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-3-4b-it/ - Context window: 0 - Parameters: 4.3B - Licence: gemma (https://ai.google.dev/gemma/terms) - Open weights: yes - Release date: 2025-02-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 3 4B It is an open-weight vision-language model from Google DeepMind with 4.3B parameters, published under gemma. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-3-4b-it, https://huggingface.co/api/models/google/gemma-3-4b-it?blobs=true, https://ai.google.dev/gemma/terms ## Gemma 4 12B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-12b-it/ - Context window: 0 - Parameters: 12B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-05-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Gemma 4 12B It is published by Google DeepMind as open weights with 12B parameters under Apache-2.0. It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The repository was created in May 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-4-12B-it, https://huggingface.co/api/models/google/gemma-4-12B-it?blobs=true, https://ai.google.dev/gemma/docs/gemma_4_license ## Gemma 4 26B A4B IT - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-26b-a4b-it/ - Context window: 262144 - Parameters: 25.81B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-03-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 4 26B A4B IT is a mid-size Gemma 4 model stored as 25.8B parameters, sitting between the 12B and 31B sizes in the family. It has the same 256K-token context window as the 31B model and takes text and image input. The model card reports 82.6% on MMLU Pro and 88.3% on AIME 2026 without tools. Sources: https://huggingface.co/google/gemma-4-26B-A4B-it ## Gemma 4 31B - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-31b/ - Context window: 0 - Parameters: 32.7B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-03-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 4 31B is published by Google DeepMind as open weights with 32.7B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-4-31B, https://huggingface.co/api/models/google/gemma-4-31B?blobs=true, https://ai.google.dev/gemma/docs/gemma_4_license ## Gemma 4 31B IT - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-31b-it/ - Context window: 262144 - Parameters: 31.27B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-03-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 4 31B (gemma-4-31b-it) is the largest instruction-tuned Gemma 4 model, at 31.3B parameters under Apache-2.0. It takes text and images over a 256K-token context window, and the card reports 85.2% on MMLU Pro, 84.3% on GPQA Diamond and 88.4% on MMMLU, all vendor-reported; serving it at BF16 needs a large-memory accelerator. Sources: https://huggingface.co/google/gemma-4-31B-it, https://ai.google.dev/gemma/docs/gemma_4_license ## Gemma 4 E2B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-e2b-it/ - Context window: 0 - Parameters: 5.1B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-03-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Gemma 4 E2B It is published by Google DeepMind as open weights with 5.1B parameters under Apache-2.0. It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The repository was created in March 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/gemma-4-E2B-it, https://huggingface.co/api/models/google/gemma-4-E2B-it?blobs=true, https://ai.google.dev/gemma/docs/gemma_4_license ## Gemma 4 E4B IT - Provider: Google DeepMind - Page: https://llmeu.com/en/models/gemma-4-e4b-it/ - Context window: 131072 - Parameters: 8B - Licence: apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license) - Open weights: yes - Release date: 2026-03-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gemma 4 E4B IT is the smallest Gemma 4 model with native audio, handling text, images and audio input with a 128K-token context window. Google states that the "E" in E2B and E4B stands for effective parameters and that these sizes use Per-Layer Embeddings, so the stored model is about 8B parameters while the effective count is smaller. The model card reports 69.4% on MMLU Pro and 58.6% on GPQA Diamond, and it is Apache-2.0 licensed. Sources: https://huggingface.co/google/gemma-4-E4B-it ## Glm 4 9B Chat - Provider: Z.ai (Zhipu AI) - Page: https://llmeu.com/en/models/glm-4-9b-chat/ - Context window: 0 - Parameters: 9.4B - Licence: other (https://huggingface.co/THUDM/glm-4-9b-chat/blob/main/LICENSE) - Open weights: yes - Release date: 2024-06-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Glm 4 9B Chat is an open-weight language model from Z.ai (Zhipu AI) with 9.4B parameters, published under glm-4 (custom, tagged "other"). The repository was created in June 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/zai-org/glm-4-9b-chat, https://huggingface.co/api/models/zai-org/glm-4-9b-chat?blobs=true, https://huggingface.co/THUDM/glm-4-9b-chat/blob/main/LICENSE ## Gpt Neo 1.3B - Provider: EleutherAI - Page: https://llmeu.com/en/models/gpt-neo-1-3b/ - Context window: 0 - Parameters: 1.4B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2022-03-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gpt Neo 1.3B is an open-weight language model from EleutherAI with 1.4B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/gpt-neo-1.3B, https://huggingface.co/api/models/EleutherAI/gpt-neo-1.3B?blobs=true, https://opensource.org/license/mit ## Gpt Neo 125M - Provider: EleutherAI - Page: https://llmeu.com/en/models/gpt-neo-125m/ - Context window: 0 - Parameters: 0.2B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2022-03-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Gpt Neo 125M is an open-weight language model from EleutherAI with 0.2B parameters, published under MIT. The published configuration records no context window, so this card carries none. The repository was created in March 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/gpt-neo-125m, https://huggingface.co/api/models/EleutherAI/gpt-neo-125m?blobs=true, https://opensource.org/license/mit ## Gpt Neo 2.7B - Provider: EleutherAI - Page: https://llmeu.com/en/models/gpt-neo-2-7b/ - Context window: 0 - Parameters: 2.7B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2022-03-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Gpt Neo 2.7B is an open-weight language model from EleutherAI with 2.7B parameters, published under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in March 2022 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/gpt-neo-2.7B, https://huggingface.co/api/models/EleutherAI/gpt-neo-2.7B?blobs=true, https://opensource.org/license/mit ## Gpt Neox 20B - Provider: EleutherAI - Page: https://llmeu.com/en/models/gpt-neox-20b/ - Context window: 0 - Parameters: 20.7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-04-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Gpt Neox 20B is published by EleutherAI as open weights with 20.7B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2022 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/gpt-neox-20b, https://huggingface.co/api/models/EleutherAI/gpt-neox-20b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Gpt Oss Safeguard 20B - Provider: OpenAI - Page: https://llmeu.com/en/models/gpt-oss-safeguard-20b/ - Context window: 0 - Parameters: 21.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Gpt Oss Safeguard 20B is a large open-weight language model from OpenAI: 21.5B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/openai/gpt-oss-safeguard-20b, https://huggingface.co/api/models/openai/gpt-oss-safeguard-20b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 3.0 1B A400M Instruct - Provider: IBM - Page: https://llmeu.com/en/models/granite-3-0-1b-a400m-instruct/ - Context window: 0 - Parameters: 1.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-10-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 3.0 1B A400M Instruct is published by IBM as open weights with 1.3B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in October 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-3.0-1b-a400m-instruct, https://huggingface.co/api/models/ibm-granite/granite-3.0-1b-a400m-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 3.1 8B Instruct - Provider: IBM - Page: https://llmeu.com/en/models/granite-3-1-8b-instruct/ - Context window: 0 - Parameters: 8.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-12-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Granite 3.1 8B Instruct is a mid-size open-weight language model from IBM: 8.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-3.1-8b-instruct, https://huggingface.co/api/models/ibm-granite/granite-3.1-8b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 3.2 8B Instruct - Provider: IBM - Page: https://llmeu.com/en/models/granite-3-2-8b-instruct/ - Context window: 0 - Parameters: 8.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-02-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 3.2 8B Instruct is an open-weight language model from IBM with 8.2B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in February 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-3.2-8b-instruct, https://huggingface.co/api/models/ibm-granite/granite-3.2-8b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 3.3 2B Instruct - Provider: IBM - Page: https://llmeu.com/en/models/granite-3-3-2b-instruct/ - Context window: 0 - Parameters: 2.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-04-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 3.3 2B Instruct is published by IBM as open weights with 2.5B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-3.3-2b-instruct, https://huggingface.co/api/models/ibm-granite/granite-3.3-2b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 3.3 8B Instruct - Provider: IBM - Page: https://llmeu.com/en/models/granite-3-3-8b-instruct/ - Context window: 0 - Parameters: 8.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-04-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Granite 3.3 8B Instruct is published by IBM as open weights with 8.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-3.3-8b-instruct, https://huggingface.co/api/models/ibm-granite/granite-3.3-8b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.0 3B Vision - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-0-3b-vision/ - Context window: 0 - Parameters: 4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-03-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Granite 4.0 3B Vision is a mid-size open-weight vision-language model from IBM: 4B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in March 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-4.0-3b-vision, https://huggingface.co/api/models/ibm-granite/granite-4.0-3b-vision?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.0 H Tiny - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-0-h-tiny/ - Context window: 0 - Parameters: 6.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 4.0 H Tiny is a mid-size open-weight language model from IBM: 6.9B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-4.0-h-tiny, https://huggingface.co/api/models/ibm-granite/granite-4.0-h-tiny?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.0 Micro - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-0-micro/ - Context window: 0 - Parameters: 3.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Granite 4.0 Micro is a mid-size open-weight language model from IBM: 3.4B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in September 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-4.0-micro, https://huggingface.co/api/models/ibm-granite/granite-4.0-micro?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.1 30B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-1-30b/ - Context window: 0 - Parameters: 28.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-04-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 4.1 30B is an open-weight language model from IBM with 28.9B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-4.1-30b, https://huggingface.co/api/models/ibm-granite/granite-4.1-30b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.1 3B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-1-3b/ - Context window: 0 - Parameters: 3.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-04-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Granite 4.1 3B is an open-weight language model from IBM with 3.4B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-4.1-3b, https://huggingface.co/api/models/ibm-granite/granite-4.1-3b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.1 8B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-1-8b/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-04-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Granite 4.1 8B is an open-weight language model from IBM with 8.8B parameters, published under Apache-2.0. The repository was created in April 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/ibm-granite/granite-4.1-8b, https://huggingface.co/api/models/ibm-granite/granite-4.1-8b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.2 30B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-2-30b/ - Context window: 0 - Parameters: 29.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Granite 4.2 30B is published by IBM as open weights with 29.3B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 13 languages, 8 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-4.2-30b, https://huggingface.co/api/models/ibm-granite/granite-4.2-30b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.2 3B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-2-3b/ - Context window: 0 - Parameters: 3.7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Granite 4.2 3B is published by IBM as open weights with 3.7B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 13 languages, 8 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-4.2-3b, https://huggingface.co/api/models/ibm-granite/granite-4.2-3b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite 4.2 8B - Provider: IBM - Page: https://llmeu.com/en/models/granite-4-2-8b/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Granite 4.2 8B is published by IBM as open weights with 8.8B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in August 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 13 languages, 8 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-4.2-8b, https://huggingface.co/api/models/ibm-granite/granite-4.2-8b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite Docling 258M - Provider: IBM - Page: https://llmeu.com/en/models/granite-docling-258m/ - Context window: 0 - Parameters: 0.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Granite Docling 258M is a small open-weight vision-language model from IBM: 0.3B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-docling-258M, https://huggingface.co/api/models/ibm-granite/granite-docling-258M?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite Guardian 4.1 8B - Provider: IBM - Page: https://llmeu.com/en/models/granite-guardian-4-1-8b/ - Context window: 0 - Parameters: 8.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-04-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Granite Guardian 4.1 8B is published by IBM as open weights with 8.4B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-guardian-4.1-8b, https://huggingface.co/api/models/ibm-granite/granite-guardian-4.1-8b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Granite Vision 4.1 4B - Provider: IBM - Page: https://llmeu.com/en/models/granite-vision-4-1-4b/ - Context window: 0 - Parameters: 4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-04-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Granite Vision 4.1 4B is a mid-size open-weight vision-language model from IBM: 4B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/ibm-granite/granite-vision-4.1-4b, https://huggingface.co/api/models/ibm-granite/granite-vision-4.1-4b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Hunyuan A13B Instruct - Provider: Tencent - Page: https://llmeu.com/en/models/hunyuan-a13b-instruct/ - Context window: 0 - Parameters: 80.4B - Licence: other (https://github.com/Tencent-Hunyuan/Hunyuan-A13B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-06-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Hunyuan A13B Instruct is an open-weight language model from Tencent with 80.4B parameters, published under tencent-hunyuan-a13b (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tencent/Hunyuan-A13B-Instruct, https://huggingface.co/api/models/tencent/Hunyuan-A13B-Instruct?blobs=true, https://github.com/Tencent-Hunyuan/Hunyuan-A13B/blob/main/LICENSE ## Hunyuan Hy3 - Provider: Tencent - Page: https://llmeu.com/en/models/hy3/ - Context window: 0 - Parameters: 298.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-07-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Hunyuan Hy3 is published by Tencent as open weights with 298.8B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/tencent/Hy3, https://huggingface.co/api/models/tencent/Hy3?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Hunyuan Hy4 Preview - Provider: Tencent - Page: https://llmeu.com/en/models/hy4-preview/ - Context window: 0 - Parameters: 780B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Hunyuan Hy4 Preview is a very large open-weight language model from Tencent: 780B parameters under Apache-2.0. The repository was created in August 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/tencent/Hy4-preview, https://huggingface.co/api/models/tencent/Hy4-preview?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Hunyuan MT2 1.8B - Provider: Tencent - Page: https://llmeu.com/en/models/hunyuan-mt2-1-8b/ - Context window: 0 - Parameters: 1.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-05-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Hunyuan MT2 1.8B is a small open-weight language model from Tencent: 1.8B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in May 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tencent/Hy-MT2-1.8B-FP8, https://huggingface.co/api/models/tencent/Hy-MT2-1.8B-FP8?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Hunyuan MT2 7B - Provider: Tencent - Page: https://llmeu.com/en/models/hunyuan-mt2-7b/ - Context window: 0 - Parameters: 7.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-05-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Hunyuan MT2 7B is a mid-size open-weight language model from Tencent: 7.5B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in May 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tencent/Hy-MT2-7B-FP8, https://huggingface.co/api/models/tencent/Hy-MT2-7B-FP8?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## HunyuanOCR - Provider: Tencent - Page: https://llmeu.com/en/models/hunyuanocr/ - Context window: 0 - Parameters: 1.1B - Licence: other (https://huggingface.co/tencent/HunyuanOCR/blob/main/LICENSE) - Open weights: yes - Release date: 2025-11-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report HunyuanOCR is published by Tencent as open weights with 1.1B parameters under tencent-hunyuan-community (custom, tagged "other"). It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tencent/HunyuanOCR, https://huggingface.co/api/models/tencent/HunyuanOCR?blobs=true, https://huggingface.co/tencent/HunyuanOCR/blob/main/LICENSE ## Instructblip Flan T5 Xl - Provider: Salesforce - Page: https://llmeu.com/en/models/instructblip-flan-t5-xl/ - Context window: 0 - Parameters: 4B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2023-05-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Instructblip Flan T5 Xl is a mid-size open-weight vision-language model from Salesforce: 4B parameters under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in May 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Salesforce/instructblip-flan-t5-xl, https://huggingface.co/api/models/Salesforce/instructblip-flan-t5-xl?blobs=true, https://opensource.org/license/mit ## Instructblip Vicuna 7B - Provider: Salesforce - Page: https://llmeu.com/en/models/instructblip-vicuna-7b/ - Context window: 0 - Parameters: 7.9B - Licence: other (https://huggingface.co/Salesforce/instructblip-vicuna-7b/blob/main/LICENSE) - Open weights: yes - Release date: 2023-05-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Instructblip Vicuna 7B is a mid-size open-weight vision-language model from Salesforce: 7.9B parameters under other. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in May 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Salesforce/instructblip-vicuna-7b, https://huggingface.co/api/models/Salesforce/instructblip-vicuna-7b?blobs=true, https://huggingface.co/Salesforce/instructblip-vicuna-7b/blob/main/LICENSE ## Intern-S1 - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/intern-s1/ - Context window: 0 - Parameters: 240.7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-07-24 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Intern-S1 is an open-weight vision-language model from Shanghai AI Laboratory with 240.7B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in July 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/Intern-S1, https://huggingface.co/api/models/internlm/Intern-S1?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Intern-S1 Mini - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/intern-s1-mini/ - Context window: 0 - Parameters: 8.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-08-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Intern-S1 Mini is an open-weight vision-language model from Shanghai AI Laboratory with 8.5B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in August 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/Intern-S1-mini, https://huggingface.co/api/models/internlm/Intern-S1-mini?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## InternLM 2 20B Chat - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/internlm2-20b-chat/ - Context window: 0 - Parameters: 19.9B - Licence: other (https://huggingface.co/internlm/internlm2-chat-20b/blob/main/LICENSE) - Open weights: yes - Release date: 2024-01-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. InternLM 2 20B Chat is published by Shanghai AI Laboratory as open weights with 19.9B parameters under other. The published configuration records no context window, so this card carries none. The repository was created in January 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/internlm2-chat-20b, https://huggingface.co/api/models/internlm/internlm2-chat-20b?blobs=true, https://huggingface.co/internlm/internlm2-chat-20b/blob/main/LICENSE ## InternLM 2 7B Chat - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/internlm2-7b-chat/ - Context window: 0 - Parameters: 7.7B - Licence: other (https://huggingface.co/internlm/internlm2-chat-7b/blob/main/LICENSE) - Open weights: yes - Release date: 2024-01-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report InternLM 2 7B Chat is published by Shanghai AI Laboratory as open weights with 7.7B parameters under other. The repository was created in January 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/internlm2-chat-7b, https://huggingface.co/api/models/internlm/internlm2-chat-7b?blobs=true, https://huggingface.co/internlm/internlm2-chat-7b/blob/main/LICENSE ## InternLM 2.5 7B Chat - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/internlm2-5-7b-chat/ - Context window: 0 - Parameters: 7.7B - Licence: other (https://huggingface.co/internlm/internlm2_5-7b-chat/blob/main/LICENSE) - Open weights: yes - Release date: 2024-06-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report InternLM 2.5 7B Chat is a mid-size open-weight language model from Shanghai AI Laboratory: 7.7B parameters under other. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/internlm2_5-7b-chat, https://huggingface.co/api/models/internlm/internlm2_5-7b-chat?blobs=true, https://huggingface.co/internlm/internlm2_5-7b-chat/blob/main/LICENSE ## InternLM 3 8B Instruct - Provider: Shanghai AI Laboratory - Page: https://llmeu.com/en/models/internlm3-8b-instruct/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-01-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. InternLM 3 8B Instruct is a mid-size open-weight language model from Shanghai AI Laboratory: 8.8B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in January 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/internlm/internlm3-8b-instruct, https://huggingface.co/api/models/internlm/internlm3-8b-instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Kimi K2 Instruct - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-instruct/ - Context window: 0 - Parameters: 1026.4B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2025-07-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Kimi K2 Instruct is published by Moonshot AI as open weights with 1026.4B parameters under modified-mit (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2-Instruct, https://huggingface.co/api/models/moonshotai/Kimi-K2-Instruct?blobs=true, https://huggingface.co/moonshotai/Kimi-K2-Instruct/blob/main/LICENSE ## Kimi K2 Instruct 0905 - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-instruct-0905/ - Context window: 0 - Parameters: 1026.5B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905/blob/main/LICENSE) - Open weights: yes - Release date: 2025-09-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Kimi K2 Instruct 0905 is published by Moonshot AI as open weights with 1026.5B parameters under modified-mit (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in September 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905, https://huggingface.co/api/models/moonshotai/Kimi-K2-Instruct-0905?blobs=true, https://huggingface.co/moonshotai/Kimi-K2-Instruct-0905/blob/main/LICENSE ## Kimi K2 Thinking - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-thinking/ - Context window: 0 - Parameters: 1026.4B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2-Thinking/blob/main/LICENSE) - Open weights: yes - Release date: 2025-11-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Kimi K2 Thinking is published by Moonshot AI as open weights with 1026.4B parameters under modified-mit (custom, tagged "other"). The published configuration records no context window, so this card carries none. The repository was created in November 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2-Thinking, https://huggingface.co/api/models/moonshotai/Kimi-K2-Thinking?blobs=true, https://huggingface.co/moonshotai/Kimi-K2-Thinking/blob/main/LICENSE ## Kimi K2.5 - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-5/ - Context window: 0 - Parameters: 1026.9B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2.5/blob/main/LICENSE) - Open weights: yes - Release date: 2026-01-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Kimi K2.5 is published by Moonshot AI as open weights with 1026.9B parameters under modified-mit (custom, tagged "other"). The published configuration records no context window, so this card carries none. The repository was created in January 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2.5, https://huggingface.co/api/models/moonshotai/Kimi-K2.5?blobs=true, https://huggingface.co/moonshotai/Kimi-K2.5/blob/main/LICENSE ## Kimi K2.6 - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-6/ - Context window: 0 - Parameters: 1026.9B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2.6/blob/main/LICENSE) - Open weights: yes - Release date: 2026-04-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Kimi K2.6 is a very large open-weight vision-language model from Moonshot AI: 1026.9B parameters under modified-mit (custom, tagged "other"). It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2.6, https://huggingface.co/api/models/moonshotai/Kimi-K2.6?blobs=true, https://huggingface.co/moonshotai/Kimi-K2.6/blob/main/LICENSE ## Kimi K2.7 Code - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k2-7-code/ - Context window: 0 - Parameters: 1026.9B - Licence: other (https://huggingface.co/moonshotai/Kimi-K2.7-Code/blob/main/LICENSE) - Open weights: yes - Release date: 2026-06-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Kimi K2.7 Code is an open-weight vision-language model from Moonshot AI with 1026.9B parameters, published under modified-mit (custom, tagged "other"). It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K2.7-Code, https://huggingface.co/api/models/moonshotai/Kimi-K2.7-Code?blobs=true, https://huggingface.co/moonshotai/Kimi-K2.7-Code/blob/main/LICENSE ## Kimi K3 - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-k3/ - Context window: 0 - Parameters: 2779.9B - Licence: other (https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE) - Open weights: yes - Release date: 2026-06-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Kimi K3 is a very large open-weight vision-language model from Moonshot AI: 2779.9B parameters under kimi-k3 (custom, tagged "other"). The repository was created in June 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-K3, https://huggingface.co/api/models/moonshotai/Kimi-K3?blobs=true, https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE ## Kimi Linear 48B A3B Instruct - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-linear-48b-a3b-instruct/ - Context window: 0 - Parameters: 49.1B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-10-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Kimi Linear 48B A3B Instruct is a large open-weight language model from Moonshot AI: 49.1B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-Linear-48B-A3B-Instruct, https://huggingface.co/api/models/moonshotai/Kimi-Linear-48B-A3B-Instruct?blobs=true, https://opensource.org/license/mit ## Kimi VL A3B Instruct - Provider: Moonshot AI - Page: https://llmeu.com/en/models/kimi-vl-a3b-instruct/ - Context window: 0 - Parameters: 16.4B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-04-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Kimi VL A3B Instruct is a large open-weight vision-language model from Moonshot AI: 16.4B parameters under MIT. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Kimi-VL-A3B-Instruct, https://huggingface.co/api/models/moonshotai/Kimi-VL-A3B-Instruct?blobs=true, https://opensource.org/license/mit ## LLMEU V2 (announced) - Provider: LLM EU - Page: https://llmeu.com/en/models/llmeu-v2/ - Context window: unverified - Parameters: unverified - Licence: unknown - Open weights: no - Release date: unverified - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report LLMEU V2 is the model LLM EU intends to train and serve on European infrastructure, aimed at the gap the current generation leaves open: the EU's smaller official languages, where coverage is thin and quality drops fastest. Nothing about it is built yet. There are no weights, no parameter count, no context window and no benchmark result, and this card will stay empty of them until there is something real to publish. It is listed for one reason: a catalogue that cannot tell an announced model from an available one is not a catalogue, it is a brochure. Sources: https://llmeu.com/en/docs/routing/ ## Llama 2 13B Chat Hf - Provider: Meta - Page: https://llmeu.com/en/models/llama-2-13b-chat-hf/ - Context window: 0 - Parameters: 13B - Licence: llama2 (https://www.llama.com/llama2/license/) - Open weights: yes - Release date: 2023-07-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Llama 2 13B Chat Hf is published by Meta as open weights with 13B parameters under llama2. The published configuration records no context window, so this card carries none. The repository was created in July 2023 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-2-13b-chat-hf, https://huggingface.co/api/models/meta-llama/Llama-2-13b-chat-hf?blobs=true, https://www.llama.com/llama2/license/ ## Llama 2 7B Chat Hf - Provider: Meta - Page: https://llmeu.com/en/models/llama-2-7b-chat-hf/ - Context window: 0 - Parameters: 6.7B - Licence: llama2 (https://www.llama.com/llama2/license/) - Open weights: yes - Release date: 2023-07-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 2 7B Chat Hf is published by Meta as open weights with 6.7B parameters under llama2. The repository was created in July 2023 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-2-7b-chat-hf, https://huggingface.co/api/models/meta-llama/Llama-2-7b-chat-hf?blobs=true, https://www.llama.com/llama2/license/ ## Llama 3.1 405B - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-1-405b/ - Context window: 0 - Parameters: 405.9B - Licence: llama3.1 (https://www.llama.com/llama3_1/license/) - Open weights: yes - Release date: 2024-07-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 3.1 405B is a very large open-weight language model from Meta: 405.9B parameters under llama3.1. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 8 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-3.1-405B, https://huggingface.co/api/models/meta-llama/Llama-3.1-405B?blobs=true, https://www.llama.com/llama3_1/license/ ## Llama 3.1 70B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-1-70b-instruct/ - Context window: 0 - Parameters: 70.6B - Licence: llama3.1 (https://www.llama.com/llama3_1/license/) - Open weights: yes - Release date: 2024-07-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 3.1 70B Instruct is a large open-weight language model from Meta: 70.6B parameters under llama3.1. The repository was created in July 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 8 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct, https://huggingface.co/api/models/meta-llama/Llama-3.1-70B-Instruct?blobs=true, https://www.llama.com/llama3_1/license/ ## Llama 3.1 8B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-1-8b-instruct/ - Context window: 131072 - Parameters: 8.03B - Licence: llama3.1 (https://www.llama.com/llama3_1/license/) - Open weights: yes - Release date: 2024-07-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 3.1 8B Instruct is Meta’s mid-small instruction model with a 128k-token context window and tool-calling support, released under the Llama 3.1 Community License. That licence permits commercial use but requires a licence from Meta above 700 million monthly active users and adds an acceptable-use policy. It remains one of the most downloaded open checkpoints and a standard baseline; in the OpenGPT-X European-language evaluation it scores 0.563 on EU21. Sources: https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct, https://www.llama.com/llama3_1/license/ ## Llama 3.2 11B Vision Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-2-11b-vision-instruct/ - Context window: 0 - Parameters: 10.7B - Licence: llama3.2 (https://www.llama.com/llama3_2/license/) - Open weights: yes - Release date: 2024-09-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Llama 3.2 11B Vision Instruct is an open-weight vision-language model from Meta with 10.7B parameters, published under llama3.2. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in September 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 8 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct, https://huggingface.co/api/models/meta-llama/Llama-3.2-11B-Vision-Instruct?blobs=true, https://www.llama.com/llama3_2/license/ ## Llama 3.2 1B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-2-1b-instruct/ - Context window: 0 - Parameters: 1.2B - Licence: llama3.2 (https://www.llama.com/llama3_2/license/) - Open weights: yes - Release date: 2024-09-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 3.2 1B Instruct is an open-weight language model from Meta with 1.2B parameters, published under llama3.2. The repository was created in September 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 8 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct, https://huggingface.co/api/models/meta-llama/Llama-3.2-1B-Instruct?blobs=true, https://www.llama.com/llama3_2/license/ ## Llama 3.2 3B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-2-3b-instruct/ - Context window: 0 - Parameters: 3.2B - Licence: llama3.2 (https://www.llama.com/llama3_2/license/) - Open weights: yes - Release date: 2024-09-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Llama 3.2 3B Instruct is an open-weight language model from Meta with 3.2B parameters, published under llama3.2. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in September 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 8 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct, https://huggingface.co/api/models/meta-llama/Llama-3.2-3B-Instruct?blobs=true, https://www.llama.com/llama3_2/license/ ## Llama 3.3 70B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-3-3-70b-instruct/ - Context window: 131072 - Parameters: 70.55B - Licence: llama3.3 (https://www.llama.com/llama3_3/license/) - Open weights: yes - Release date: 2024-12-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 3.3 70B Instruct is the December 2024 refresh of Meta’s 70B instruct model, with a 128k-token context window and the same Llama 3.3 Community License terms as Llama 3.1. Meta positions it as an alternative to the larger Llama 3.1 405B for general assistant work. It is text-only and is the largest dense Llama checkpoint still in common self-hosted use. Sources: https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct, https://www.llama.com/llama3_3/license/ ## Llama 4 Maverick 17B-128E Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-4-maverick-17b-128e-instruct/ - Context window: 1048576 - Parameters: 401.58B - Licence: other (https://www.llama.com/llama4/license/) - Open weights: yes - Release date: 2025-04-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 4 Maverick is Meta’s larger Llama 4 mixture-of-experts model: 17B active parameters, 128 experts and about 400B total parameters, with a published context window of 1M tokens. It takes text and images and is positioned for general assistant and image-understanding work. Like Scout it ships under the Llama 4 Community License with the 700M monthly-active-user threshold. Sources: https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct, https://ai.meta.com/blog/llama-4-multimodal-intelligence/, https://docs.aws.eu/bedrock/latest/userguide/model-card-meta-llama-4-maverick-17b-instruct.html ## Llama 4 Scout 17B-16E Instruct - Provider: Meta - Page: https://llmeu.com/en/models/llama-4-scout-17b-16e-instruct/ - Context window: 10485760 - Parameters: 108.64B - Licence: other (https://www.llama.com/llama4/license/) - Open weights: yes - Release date: 2025-04-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama 4 Scout is a natively multimodal mixture-of-experts model with 17B active parameters, 16 experts and 109B total parameters. Meta advertises a 10 million token context window for it, the largest of any Llama release, and states it fits on a single H100 with Int4 quantisation. It is released under the Llama 4 Community License, which keeps the 700M monthly-active-user threshold. Sources: https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct, https://ai.meta.com/blog/llama-4-multimodal-intelligence/ ## Llama Embed Nemotron 8B - Provider: NVIDIA - Page: https://llmeu.com/en/models/llama-embed-nemotron-8b/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/nvidia/llama-embed-nemotron-8b/blob/main/LICENSE) - Open weights: yes - Release date: 2025-10-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama Embed Nemotron 8B is published by NVIDIA as open weights with an unpublished parameter count under other. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/nvidia/llama-embed-nemotron-8b, https://huggingface.co/api/models/nvidia/llama-embed-nemotron-8b?blobs=true, https://huggingface.co/nvidia/llama-embed-nemotron-8b/blob/main/LICENSE ## Llama Guard 3 8B - Provider: Meta - Page: https://llmeu.com/en/models/llama-guard-3-8b/ - Context window: 0 - Parameters: 8B - Licence: llama3.1 (https://www.llama.com/llama3_1/license/) - Open weights: yes - Release date: 2024-07-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Llama Guard 3 8B is an open-weight language model from Meta with 8B parameters, published under llama3.1. The published configuration records no context window, so this card carries none. The repository was created in July 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-Guard-3-8B, https://huggingface.co/api/models/meta-llama/Llama-Guard-3-8B?blobs=true, https://www.llama.com/llama3_1/license/ ## Llama Guard 4 12B - Provider: Meta - Page: https://llmeu.com/en/models/llama-guard-4-12b/ - Context window: 0 - Parameters: 12B - Licence: other (https://huggingface.co/meta-llama/Llama-Guard-4-12B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Llama Guard 4 12B is an open-weight vision-language model from Meta with 12B parameters, published under llama4 (custom, tagged "other"). The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Llama-Guard-4-12B, https://huggingface.co/api/models/meta-llama/Llama-Guard-4-12B?blobs=true, https://huggingface.co/meta-llama/Llama-Guard-4-12B/blob/main/LICENSE ## Llama Nemotron Embed 1B V2 - Provider: NVIDIA - Page: https://llmeu.com/en/models/llama-nemotron-embed-1b-v2/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/nvidia/llama-nemotron-embed-1b-v2/blob/main/LICENSE) - Open weights: yes - Release date: 2025-10-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It turns text into vectors for retrieval instead of generating answers. Llama Nemotron Embed 1B V2 is published by NVIDIA as open weights with an unpublished parameter count under other. The published configuration records no context window, so this card carries none. The repository was created in October 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/nvidia/llama-nemotron-embed-1b-v2, https://huggingface.co/api/models/nvidia/llama-nemotron-embed-1b-v2?blobs=true, https://huggingface.co/nvidia/llama-nemotron-embed-1b-v2/blob/main/LICENSE ## Llama Nemotron Rerank 1B v2 - Provider: NVIDIA - Page: https://llmeu.com/en/models/llama-nemotron-rerank-1b-v2/ - Context window: 8192 - Parameters: 1.24B - Licence: other (https://openmdw.ai/license/1-1/) - Open weights: yes - Release date: 2025-10-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Llama Nemotron Rerank 1B v2 is a cross-encoder reranker released under the OpenMDW-1.1 licence. The model card states it was evaluated on 26 languages and works up to a sequence length of 8192 tokens, with longer texts to be chunked or truncated. It is intended for second-stage ranking after a vector search in a retrieval pipeline. Sources: https://huggingface.co/nvidia/llama-nemotron-rerank-1b-v2 ## Llama Nemotron Rerank Vl 1B V2 - Provider: NVIDIA - Page: https://llmeu.com/en/models/llama-nemotron-rerank-vl-1b-v2/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/nvidia/llama-nemotron-rerank-vl-1b-v2/blob/main/LICENSE) - Open weights: yes - Release date: 2025-12-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Llama Nemotron Rerank Vl 1B V2 is published by NVIDIA as open weights with an unpublished parameter count under other. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in December 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/nvidia/llama-nemotron-rerank-vl-1b-v2, https://huggingface.co/api/models/nvidia/llama-nemotron-rerank-vl-1b-v2?blobs=true, https://huggingface.co/nvidia/llama-nemotron-rerank-vl-1b-v2/blob/main/LICENSE ## Magistral Small 1.2 - Provider: Mistral AI - Page: https://llmeu.com/en/models/magistral-small-1-2/ - Context window: 131072 - Parameters: 24.01B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Magistral Small 1.2 is a 24B Apache-2.0 reasoning model derived from Mistral Small 3.2, with a vision encoder and explicit think tokens for the reasoning trace. The vendor reports AIME24 at 86.14%, AIME25 at 77.34% and GPQA Diamond at 70.07%. Mistral notes that quality may degrade beyond roughly 40k tokens even though the window is 128k. Sources: https://huggingface.co/mistralai/Magistral-Small-2509 ## Medgemma 4B It - Provider: Google DeepMind - Page: https://llmeu.com/en/models/medgemma-4b-it/ - Context window: 0 - Parameters: 4.3B - Licence: other (https://developers.google.com/health-ai-developer-foundations/terms) - Open weights: yes - Release date: 2025-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Medgemma 4B It is a mid-size open-weight vision-language model from Google DeepMind: 4.3B parameters under health-ai-developer-foundations (custom, tagged "other"). The repository was created in May 2025 and is mirrored on Hugging Face. It accepts text and images and answers in text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/google/medgemma-4b-it, https://huggingface.co/api/models/google/medgemma-4b-it?blobs=true, https://developers.google.com/health-ai-developer-foundations/terms ## Meta Llama 3 70B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/meta-llama-3-70b-instruct/ - Context window: 0 - Parameters: 70.6B - Licence: llama3 (https://www.llama.com/llama3/license/) - Open weights: yes - Release date: 2024-04-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Meta Llama 3 70B Instruct is a large open-weight language model from Meta: 70.6B parameters under llama3. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct, https://huggingface.co/api/models/meta-llama/Meta-Llama-3-70B-Instruct?blobs=true, https://www.llama.com/llama3/license/ ## Meta Llama 3 8B Instruct - Provider: Meta - Page: https://llmeu.com/en/models/meta-llama-3-8b-instruct/ - Context window: 0 - Parameters: 8B - Licence: llama3 (https://www.llama.com/llama3/license/) - Open weights: yes - Release date: 2024-04-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Meta Llama 3 8B Instruct is a mid-size open-weight language model from Meta: 8B parameters under llama3. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct, https://huggingface.co/api/models/meta-llama/Meta-Llama-3-8B-Instruct?blobs=true, https://www.llama.com/llama3/license/ ## MiniMax M1 40k - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-m1-40k/ - Context window: 0 - Parameters: 456.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. MiniMax M1 40k is a very large open-weight language model from MiniMax: 456.1B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/MiniMaxAI/MiniMax-M1-40k, https://huggingface.co/api/models/MiniMaxAI/MiniMax-M1-40k?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## MiniMax M2 - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-m2/ - Context window: 0 - Parameters: 228.7B - Licence: other (https://github.com/MiniMax-AI/MiniMax-M2/blob/main/LICENSE) - Open weights: yes - Release date: 2025-10-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report MiniMax M2 is an open-weight language model from MiniMax with 228.7B parameters, published under modified-mit (custom, tagged "other"). The repository was created in October 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/MiniMaxAI/MiniMax-M2, https://huggingface.co/api/models/MiniMaxAI/MiniMax-M2?blobs=true, https://github.com/MiniMax-AI/MiniMax-M2/blob/main/LICENSE ## MiniMax M2.1 - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-m2-1/ - Context window: 0 - Parameters: 228.7B - Licence: other (https://github.com/MiniMax-AI/MiniMax-M2.1/blob/main/LICENSE) - Open weights: yes - Release date: 2025-12-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report MiniMax M2.1 is an open-weight language model from MiniMax with 228.7B parameters, published under modified-mit (custom, tagged "other"). The repository was created in December 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/MiniMaxAI/MiniMax-M2.1, https://huggingface.co/api/models/MiniMaxAI/MiniMax-M2.1?blobs=true, https://github.com/MiniMax-AI/MiniMax-M2.1/blob/main/LICENSE ## MiniMax M2.5 - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-m2-5/ - Context window: 0 - Parameters: 228.7B - Licence: other (https://github.com/MiniMax-AI/MiniMax-M2.5/blob/main/LICENSE) - Open weights: yes - Release date: 2026-02-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report MiniMax M2.5 is an open-weight language model from MiniMax with 228.7B parameters, published under modified-mit (custom, tagged "other"). The repository was created in February 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/MiniMaxAI/MiniMax-M2.5, https://huggingface.co/api/models/MiniMaxAI/MiniMax-M2.5?blobs=true, https://github.com/MiniMax-AI/MiniMax-M2.5/blob/main/LICENSE ## MiniMax M3 - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-m3/ - Context window: 0 - Parameters: 427B - Licence: other (https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE) - Open weights: yes - Release date: 2026-06-02 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report MiniMax M3 is published by MiniMax as open weights with 427B parameters under minimax-community (custom, tagged "other"). The repository was created in June 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/MiniMaxAI/MiniMax-M3, https://huggingface.co/api/models/MiniMaxAI/MiniMax-M3?blobs=true, https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE ## MiniMax Text 01 - Provider: MiniMax - Page: https://llmeu.com/en/models/minimax-text-01/ - Context window: 0 - Parameters: 456.1B - Licence: other (https://huggingface.co/MiniMaxAI/MiniMax-Text-01-hf/blob/main/LICENSE) - Open weights: yes - Release date: 2025-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report MiniMax Text 01 is a very large open-weight language model from MiniMax: 456.1B parameters under minimax (custom, tagged "other"). The repository was created in June 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/MiniMaxAI/MiniMax-Text-01-hf, https://huggingface.co/api/models/MiniMaxAI/MiniMax-Text-01-hf?blobs=true, https://huggingface.co/MiniMaxAI/MiniMax-Text-01-hf/blob/main/LICENSE ## Ministral 3 8B - Provider: Mistral AI - Page: https://llmeu.com/en/models/ministral-3-8b/ - Context window: 262144 - Parameters: 8.92B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-10-31 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Ministral 3 8B is a small Apache-2.0 instruct model with vision input and a 256k-token context window. Mistral targets edge deployment with it: the vendor states the FP8 checkpoint fits in 12GB of VRAM and less when further quantised. Published self-reported scores put it close to Qwen3-VL-8B-Thinking on AIME and GPQA. Sources: https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512 ## Mistral 7B Instruct v0.3 - Provider: Mistral AI - Page: https://llmeu.com/en/models/mistral-7b-instruct-v0-3/ - Context window: 32768 - Parameters: 7.25B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-05-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Mistral 7B Instruct v0.3 is the extended-vocabulary version of Mistral’s original 7B instruct model, with a 32k-token context window and function-calling support. It is still one of the most downloaded open checkpoints and is frequently used as a baseline: the OpenGPT-X evaluation of European languages reports an EU21 average of 0.527 for it. It is well behind current models on reasoning and coding. Sources: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3, https://huggingface.co/openGPT-X/Teuken-7B-instruct-v0.6 ## Mistral Large 3 (675B) - Provider: Mistral AI - Page: https://llmeu.com/en/models/mistral-large-3-675b/ - Context window: 262144 - Parameters: 675B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Mistral Large 3 is the largest published Mistral model: a granular mixture-of-experts model with 675B total parameters and 41B active parameters, released under Apache-2.0. It is multimodal and supports a 256k-token context window. The vendor states it was trained from scratch on 3000 H200 GPUs. Sources: https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512 ## Mistral NeMo 12B Instruct - Provider: Mistral AI - Page: https://llmeu.com/en/models/mistral-nemo-12b/ - Context window: 131072 - Parameters: 12.25B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-07-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Mistral NeMo is a 12B model built with NVIDIA, released under Apache-2.0 with a 128k-token context window and a Tekken tokenizer. The card reports 68.0% on 5-shot MMLU and lists English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese and Russian as supported languages. It remains a common choice where a permissive licence and a small footprint matter more than peak quality. Sources: https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407 ## Mistral Small 4 (119B-A6B) - Provider: Mistral AI - Page: https://llmeu.com/en/models/mistral-small-4-119b-a6b/ - Context window: 262144 - Parameters: 119.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Mistral Small 4 (mistral-small-4-119b-a6b) is an Apache-2.0 mixture of experts with 119B total parameters and 6.5B activated per token. It takes text and images over a 256k-token context window, the card records one checkpoint for instruct, reasoning and coding use plus an NVFP4 checkpoint for smaller servers, and the full BF16 weights need a large-memory accelerator. Sources: https://huggingface.co/mistralai/Mistral-Small-4-119B-2603, https://docs.mistral.ai/models ## Molmo 7B D 0924 - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/molmo-7b-d-0924/ - Context window: 0 - Parameters: 8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-09-25 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Molmo 7B D 0924 is an open-weight vision-language model from Allen Institute for AI (Ai2) with 8B parameters, published under Apache-2.0. The repository was created in September 2024 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Molmo-7B-D-0924, https://huggingface.co/api/models/allenai/Molmo-7B-D-0924?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Molmo2 4B - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/molmo2-4b/ - Context window: 0 - Parameters: 4.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-12-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Molmo2 4B is a mid-size open-weight vision-language model from Allen Institute for AI (Ai2): 4.9B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in December 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Molmo2-4B, https://huggingface.co/api/models/allenai/Molmo2-4B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Molmo2 8B - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/molmo2-8b/ - Context window: 0 - Parameters: 8.7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-12-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Molmo2 8B is a mid-size open-weight vision-language model from Allen Institute for AI (Ai2): 8.7B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in December 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Molmo2-8B, https://huggingface.co/api/models/allenai/Molmo2-8B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Moonlight 16B A3B Instruct - Provider: Moonshot AI - Page: https://llmeu.com/en/models/moonlight-16b-a3b-instruct/ - Context window: 0 - Parameters: 16B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-02-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Moonlight 16B A3B Instruct is published by Moonshot AI as open weights with 16B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in February 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/moonshotai/Moonlight-16B-A3B-Instruct, https://huggingface.co/api/models/moonshotai/Moonlight-16B-A3B-Instruct?blobs=true, https://opensource.org/license/mit ## NVIDIA Canary 1B v2 - Provider: NVIDIA - Page: https://llmeu.com/en/models/canary-1b-v2/ - Context window: 0 - Parameters: unverified - Licence: cc-by-4.0 (https://creativecommons.org/licenses/by/4.0/) - Open weights: yes - Release date: 2025-08-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. NVIDIA Canary 1B v2 is an open-weight speech model from NVIDIA with an unpublished parameter count, published under cc-by-4.0. It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The repository was created in August 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 25 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/canary-1b-v2, https://huggingface.co/api/models/nvidia/canary-1b-v2?blobs=true, https://creativecommons.org/licenses/by/4.0/ ## NVIDIA Canary Qwen 2.5B - Provider: NVIDIA - Page: https://llmeu.com/en/models/canary-qwen-2-5b/ - Context window: 0 - Parameters: unverified - Licence: cc-by-4.0 (https://creativecommons.org/licenses/by/4.0/) - Open weights: yes - Release date: 2025-06-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Canary Qwen 2.5B is an open-weight speech model from NVIDIA with an unpublished parameter count, published under cc-by-4.0. The repository was created in June 2025 and is mirrored on Hugging Face. It takes audio in and writes text out. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/canary-qwen-2.5b, https://huggingface.co/api/models/nvidia/canary-qwen-2.5b?blobs=true, https://creativecommons.org/licenses/by/4.0/ ## NVIDIA Cosmos 3 Edge - Provider: NVIDIA - Page: https://llmeu.com/en/models/cosmos-3-edge/ - Context window: 0 - Parameters: 3.9B - Licence: other (https://openmdw.ai/license/1-1/) - Open weights: yes - Release date: 2026-07-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Cosmos 3 Edge is published by NVIDIA as open weights with 3.9B parameters under openmdw1.1-license (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/nvidia/Cosmos3-Edge, https://huggingface.co/api/models/nvidia/Cosmos3-Edge?blobs=true, https://openmdw.ai/license/1-1/ ## NVIDIA Cosmos Reason2 2B - Provider: NVIDIA - Page: https://llmeu.com/en/models/cosmos-reason2-2b/ - Context window: 0 - Parameters: 2.4B - Licence: other (https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license) - Open weights: yes - Release date: 2025-12-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Cosmos Reason2 2B is published by NVIDIA as open weights with 2.4B parameters under nvidia-open-model-license (custom, tagged "other"). It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/nvidia/Cosmos-Reason2-2B, https://huggingface.co/api/models/nvidia/Cosmos-Reason2-2B?blobs=true, https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license ## NVIDIA NV-Embed-v2 - Provider: NVIDIA - Page: https://llmeu.com/en/models/nv-embed-v2/ - Context window: 0 - Parameters: unverified - Licence: cc-by-nc-4.0 (https://creativecommons.org/licenses/by-nc/4.0/) - Open weights: yes - Release date: 2024-08-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA NV-Embed-v2 is published by NVIDIA as open weights with an unpublished parameter count under cc-by-nc-4.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/NV-Embed-v2, https://huggingface.co/api/models/nvidia/NV-Embed-v2?blobs=true, https://creativecommons.org/licenses/by-nc/4.0/ ## NVIDIA Nemotron 3 Embed 1B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-embed-1b/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/nvidia/Nemotron-3-Embed-1B-BF16/blob/main/LICENSE) - Open weights: yes - Release date: 2026-07-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Nemotron 3 Embed 1B is published by NVIDIA as open weights with an unpublished parameter count under other. The repository was created in July 2026 and is mirrored on Hugging Face. It turns text into vectors for retrieval instead of generating answers. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 34 languages, 12 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/Nemotron-3-Embed-1B-BF16, https://huggingface.co/api/models/nvidia/Nemotron-3-Embed-1B-BF16?blobs=true, https://huggingface.co/nvidia/Nemotron-3-Embed-1B-BF16/blob/main/LICENSE ## NVIDIA Nemotron 3 Embed 8B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-embed-8b/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/nvidia/Nemotron-3-Embed-8B-BF16/blob/main/LICENSE) - Open weights: yes - Release date: 2026-07-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It turns text into vectors for retrieval instead of generating answers. NVIDIA Nemotron 3 Embed 8B is published by NVIDIA as open weights with an unpublished parameter count under other. The published configuration records no context window, so this card carries none. The repository was created in July 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 34 languages, 12 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/Nemotron-3-Embed-8B-BF16, https://huggingface.co/api/models/nvidia/Nemotron-3-Embed-8B-BF16?blobs=true, https://huggingface.co/nvidia/Nemotron-3-Embed-8B-BF16/blob/main/LICENSE ## NVIDIA Nemotron 3 Nano 30B-A3B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-nano-30b-a3b/ - Context window: 0 - Parameters: 31.6B - Licence: other (https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/) - Open weights: yes - Release date: 2025-12-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Nemotron 3 Nano 30B-A3B is published by NVIDIA as open weights with 31.6B parameters under nvidia-nemotron-open-model-license (custom, tagged "other"). The repository was created in December 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 6 languages, 5 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16, https://huggingface.co/api/models/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16?blobs=true, https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/ ## NVIDIA Nemotron 3 Nano 4B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-nano-4b/ - Context window: 262144 - Parameters: 3.97B - Licence: other (https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/) - Open weights: yes - Release date: 2026-03-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Nemotron 3 Nano 4B is a 4B-parameter model with a context window of up to 262K tokens, released under the NVIDIA Nemotron Open Model License. Published scores include 91.1 on RULER at 128k, 78.5 on AIME25 and 95.4 on MATH500. The model card lists English as the supported language, so it is not a multilingual option. Sources: https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 ## NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-nano-omni-30b-a3b-reasoning/ - Context window: 0 - Parameters: 33B - Licence: other (https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-agreement/) - Open weights: yes - Release date: 2026-04-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning is published by NVIDIA as open weights with 33B parameters under nvidia-open-model-agreement (custom, tagged "other"). It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16, https://huggingface.co/api/models/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16?blobs=true, https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-agreement/ ## NVIDIA Nemotron 3 Super 120B-A12B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-super-120b-a12b/ - Context window: 262144 - Parameters: 123.61B - Licence: other (https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-nemotron-open-model-license/) - Open weights: yes - Release date: 2026-03-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Nemotron 3 Super is NVIDIA’s 123B mixture-of-experts model with about 12B active parameters, released under the NVIDIA Nemotron Open Model License. Its configuration allows 262,144 tokens and the model card advertises context up to 1M tokens, with RULER scores of 96.30 at 256k and 91.75 at 1M. The card lists English, French, Spanish, Italian, German, Japanese and Chinese as supported languages. Sources: https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 ## NVIDIA Nemotron 3 Ultra 550B-A55B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-ultra-550b-a55b/ - Context window: 0 - Parameters: 560.5B - Licence: other (https://openmdw.ai/license/1-1/) - Open weights: yes - Release date: 2026-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Nemotron 3 Ultra 550B-A55B is published by NVIDIA as open weights with 560.5B parameters under openmdw-1.1 (custom, tagged "other"). The repository was created in June 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 13 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16, https://huggingface.co/api/models/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16?blobs=true, https://openmdw.ai/license/1-1/ ## NVIDIA Nemotron 3.5 Lightning 30B-A3B - Provider: NVIDIA - Page: https://llmeu.com/en/models/nemotron-3-5-lightning-30b-a3b/ - Context window: 0 - Parameters: 31.6B - Licence: other (https://openmdw.ai/license/1-1/) - Open weights: yes - Release date: 2026-08-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report NVIDIA Nemotron 3.5 Lightning 30B-A3B is a large open-weight language model from NVIDIA: 31.6B parameters under openmdw-1.1 (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 6 languages, 5 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, https://huggingface.co/api/models/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16?blobs=true, https://openmdw.ai/license/1-1/ ## NVIDIA Parakeet TDT 0.6B v3 - Provider: NVIDIA - Page: https://llmeu.com/en/models/parakeet-tdt-0-6b-v3/ - Context window: 0 - Parameters: unverified - Licence: cc-by-4.0 (https://creativecommons.org/licenses/by/4.0/) - Open weights: yes - Release date: 2025-08-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. NVIDIA Parakeet TDT 0.6B v3 is published by NVIDIA as open weights with an unpublished parameter count under cc-by-4.0. It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The repository was created in August 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 25 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3, https://huggingface.co/api/models/nvidia/parakeet-tdt-0.6b-v3?blobs=true, https://creativecommons.org/licenses/by/4.0/ ## North Micro Vision Instruct - Provider: Cohere - Page: https://llmeu.com/en/models/north-micro-vision-instruct/ - Context window: 0 - Parameters: 2.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. North Micro Vision Instruct is published by Cohere as open weights with 2.5B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in August 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 11 languages, 6 of them among the languages the European Union publishes in. Sources: https://huggingface.co/CohereLabs/North-Micro-Vision-Instruct, https://huggingface.co/api/models/CohereLabs/North-Micro-Vision-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## North Mini Code 1.0 - Provider: Cohere - Page: https://llmeu.com/en/models/north-mini-code-1-0/ - Context window: 0 - Parameters: 30.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-06-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. North Mini Code 1.0 is published by Cohere as open weights with 30.5B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in June 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/CohereLabs/North-Mini-Code-1.0, https://huggingface.co/api/models/CohereLabs/North-Mini-Code-1.0?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OLMo 1B Hf - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-1b-hf/ - Context window: 0 - Parameters: 1.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-04-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report OLMo 1B Hf is published by Allen Institute for AI (Ai2) as open weights with 1.2B parameters under Apache-2.0. The repository was created in April 2024 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/OLMo-1B-hf, https://huggingface.co/api/models/allenai/OLMo-1B-hf?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OLMo 2 0325 32B Instruct - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-2-0325-32b-instruct/ - Context window: 0 - Parameters: 32.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-03-12 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report OLMo 2 0325 32B Instruct is a large open-weight language model from Allen Institute for AI (Ai2): 32.2B parameters under Apache-2.0. The repository was created in March 2025 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/OLMo-2-0325-32B-Instruct, https://huggingface.co/api/models/allenai/OLMo-2-0325-32B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OLMo 2 0425 1B Instruct - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-2-0425-1b-instruct/ - Context window: 0 - Parameters: 1.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-04-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. OLMo 2 0425 1B Instruct is a small open-weight language model from Allen Institute for AI (Ai2): 1.5B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/OLMo-2-0425-1B-Instruct, https://huggingface.co/api/models/allenai/OLMo-2-0425-1B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OLMo 2 1124 7B Instruct - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-2-1124-7b-instruct/ - Context window: 0 - Parameters: 7.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-12-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. OLMo 2 1124 7B Instruct is a mid-size open-weight language model from Allen Institute for AI (Ai2): 7.3B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in December 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/OLMo-2-1124-7B-Instruct, https://huggingface.co/api/models/allenai/OLMo-2-1124-7B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OLMoE 1B 7B 0125 Instruct - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmoe-1b-7b-0125-instruct/ - Context window: 0 - Parameters: 6.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-01-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report OLMoE 1B 7B 0125 Instruct is published by Allen Institute for AI (Ai2) as open weights with 6.9B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/OLMoE-1B-7B-0125-Instruct, https://huggingface.co/api/models/allenai/OLMoE-1B-7B-0125-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OlmOCR 2 7B 1025 - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmocr-2-7b-1025/ - Context window: 0 - Parameters: 8.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-10-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report OlmOCR 2 7B 1025 is published by Allen Institute for AI (Ai2) as open weights with 8.3B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/olmOCR-2-7B-1025, https://huggingface.co/api/models/allenai/olmOCR-2-7B-1025?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OlmOCR 7B 0225 Preview - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmocr-7b-0225-preview/ - Context window: 0 - Parameters: 8.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-01-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report OlmOCR 7B 0225 Preview is published by Allen Institute for AI (Ai2) as open weights with 8.3B parameters under Apache-2.0. The repository was created in January 2025 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/olmOCR-7B-0225-preview, https://huggingface.co/api/models/allenai/olmOCR-7B-0225-preview?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Olmo 3 1125 32B - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-3-1125-32b/ - Context window: 0 - Parameters: 32.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-04 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Olmo 3 1125 32B is an open-weight language model from Allen Institute for AI (Ai2) with 32.2B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in November 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Olmo-3-1125-32B, https://huggingface.co/api/models/allenai/Olmo-3-1125-32B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Olmo 3 32B Think SFT - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-3-32b-think-sft/ - Context window: 0 - Parameters: 32.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Olmo 3 32B Think SFT is an open-weight language model from Allen Institute for AI (Ai2) with 32.2B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in November 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Olmo-3-32B-Think-SFT, https://huggingface.co/api/models/allenai/Olmo-3-32B-Think-SFT?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Olmo 3 7B Instruct - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-3-7b-instruct/ - Context window: 0 - Parameters: 7.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Olmo 3 7B Instruct is an open-weight language model from Allen Institute for AI (Ai2) with 7.3B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in November 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Olmo-3-7B-Instruct, https://huggingface.co/api/models/allenai/Olmo-3-7B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Olmo 3 7B Think - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/olmo-3-7b-think/ - Context window: 0 - Parameters: 7.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-11-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Olmo 3 7B Think is an open-weight language model from Allen Institute for AI (Ai2) with 7.3B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in November 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/Olmo-3-7B-Think, https://huggingface.co/api/models/allenai/Olmo-3-7B-Think?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## OpenEuroLLM final models (planned) - Provider: OpenEuroLLM - Page: https://llmeu.com/en/models/openeurollm-final-release/ - Context window: unverified - Parameters: unverified - Licence: unknown - Open weights: no - Release date: unverified - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report This is a placeholder for the final OpenEuroLLM model release, scheduled in the project plan for 31 January 2028 together with the training and evaluation reports. The project has published no weights at the time this card was written. It is listed so that the catalogue distinguishes an announced model from an available one. Sources: https://openeurollm.eu/deliverables ## OpenEuroLLM initial model release (planned) - Provider: OpenEuroLLM - Page: https://llmeu.com/en/models/openeurollm-initial-release/ - Context window: unverified - Parameters: unverified - Licence: unknown - Open weights: no - Release date: unverified - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report This is a placeholder for the first OpenEuroLLM model release, which the project lists as a deliverable due on 31 December 2026. No weights, tokenizer, parameter count or context window have been published, so nothing about the model can be stated. The card exists so the platform can show the project alongside models that are already downloadable. Sources: https://openeurollm.eu/, https://openeurollm.eu/deliverables ## Pharia-1-Embedding-4608-control - Provider: Aleph Alpha - Page: https://llmeu.com/en/models/pharia-1-embedding-4608-control/ - Context window: unverified - Parameters: 7.03B - Licence: other (https://huggingface.co/Aleph-Alpha/Pharia-1-LLM-7B-control-aligned-hf) - Open weights: yes - Release date: 2024-12-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Pharia-1-Embedding-4608-control is an embedding model that Aleph Alpha built on top of Pharia-1-LLM-7B-control and published under the same Open Aleph License. The model card does not state a maximum input length, so no context window is listed here. As with the Pharia language models, commercial use is handled through Aleph Alpha’s on-premise channel. Sources: https://huggingface.co/Aleph-Alpha/Pharia-1-Embedding-4608-control-hf ## Pharia-1-LLM-7B-control-aligned - Provider: Aleph Alpha - Page: https://llmeu.com/en/models/pharia-1-llm-7b-control-aligned/ - Context window: 8192 - Parameters: 7.04B - Licence: other (https://huggingface.co/Aleph-Alpha/Pharia-1-LLM-7B-control-aligned-hf) - Open weights: yes - Release date: 2024-08-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Pharia-1-LLM-7B-control-aligned is Aleph Alpha’s 7B control model, published on Hugging Face under the Open Aleph License, which limits use to educational and research purposes. Aleph Alpha states the models were trained on curated data in compliance with EU and national copyright and data-protection rules and are optimised for German, French and Spanish. Commercial use runs through an on-premise agreement that includes the weights, not through the public download. Sources: https://huggingface.co/Aleph-Alpha/Pharia-1-LLM-7B-control, https://docs.aleph-alpha.com/phariaai-dev-guide/latest/pharia-llm/model-access.html ## Phi 1 5 - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-1-5/ - Context window: 0 - Parameters: 1.4B - Licence: mit (https://huggingface.co/microsoft/phi-1_5/resolve/main/LICENSE) - Open weights: yes - Release date: 2023-09-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Phi 1 5 is published by Microsoft as open weights with 1.4B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in September 2023 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/phi-1_5, https://huggingface.co/api/models/microsoft/phi-1_5?blobs=true, https://huggingface.co/microsoft/phi-1_5/resolve/main/LICENSE ## Phi 2 - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-2/ - Context window: 0 - Parameters: 2.8B - Licence: mit (https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE) - Open weights: yes - Release date: 2023-12-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Phi 2 is published by Microsoft as open weights with 2.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in December 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/phi-2, https://huggingface.co/api/models/microsoft/phi-2?blobs=true, https://huggingface.co/microsoft/phi-2/resolve/main/LICENSE ## Phi 3 Mini 128K Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-mini-128k-instruct/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-04-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Phi 3 Mini 128K Instruct is published by Microsoft as open weights with 3.8B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in April 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-3-mini-128k-instruct, https://huggingface.co/api/models/microsoft/Phi-3-mini-128k-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/resolve/main/LICENSE ## Phi 3 Mini 4K Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-mini-4k-instruct/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-04-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Phi 3 Mini 4K Instruct is published by Microsoft as open weights with 3.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 2 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct, https://huggingface.co/api/models/microsoft/Phi-3-mini-4k-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/resolve/main/LICENSE ## Phi 3 Vision 128K Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-vision-128k-instruct/ - Context window: 0 - Parameters: 4.1B - Licence: mit (https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-05-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Phi 3 Vision 128K Instruct is published by Microsoft as open weights with 4.1B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in May 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/microsoft/Phi-3-vision-128k-instruct, https://huggingface.co/api/models/microsoft/Phi-3-vision-128k-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3-vision-128k-instruct/resolve/main/LICENSE ## Phi 3.5 Mini Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-5-mini-instruct/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://huggingface.co/microsoft/Phi-3.5-mini-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-08-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Phi 3.5 Mini Instruct is published by Microsoft as open weights with 3.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in August 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/microsoft/Phi-3.5-mini-instruct, https://huggingface.co/api/models/microsoft/Phi-3.5-mini-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3.5-mini-instruct/resolve/main/LICENSE ## Phi 3.5 MoE Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-5-moe-instruct/ - Context window: 0 - Parameters: 41.9B - Licence: mit (https://huggingface.co/microsoft/Phi-3.5-MoE-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-08-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Phi 3.5 MoE Instruct is published by Microsoft as open weights with 41.9B parameters under MIT. The repository was created in August 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/microsoft/Phi-3.5-MoE-instruct, https://huggingface.co/api/models/microsoft/Phi-3.5-MoE-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3.5-MoE-instruct/resolve/main/LICENSE ## Phi 3.5 Vision Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-3-5-vision-instruct/ - Context window: 0 - Parameters: 4.1B - Licence: mit (https://huggingface.co/microsoft/Phi-3.5-vision-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2024-08-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Phi 3.5 Vision Instruct is published by Microsoft as open weights with 4.1B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in August 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/microsoft/Phi-3.5-vision-instruct, https://huggingface.co/api/models/microsoft/Phi-3.5-vision-instruct?blobs=true, https://huggingface.co/microsoft/Phi-3.5-vision-instruct/resolve/main/LICENSE ## Phi 4 Mini Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-4-mini-instruct/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/LICENSE) - Open weights: yes - Release date: 2025-02-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Phi 4 Mini Instruct is published by Microsoft as open weights with 3.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 24 languages, 13 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-4-mini-instruct, https://huggingface.co/api/models/microsoft/Phi-4-mini-instruct?blobs=true, https://huggingface.co/microsoft/Phi-4-mini-instruct/resolve/main/LICENSE ## Phi 4 Mini Reasoning - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-4-mini-reasoning/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://huggingface.co/microsoft/Phi-4-mini-instruct-reasoning/resolve/main/LICENSE) - Open weights: yes - Release date: 2025-04-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Phi 4 Mini Reasoning is published by Microsoft as open weights with 3.8B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-4-mini-reasoning, https://huggingface.co/api/models/microsoft/Phi-4-mini-reasoning?blobs=true, https://huggingface.co/microsoft/Phi-4-mini-instruct-reasoning/resolve/main/LICENSE ## Phi 4 Reasoning - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-4-reasoning/ - Context window: 0 - Parameters: 14.7B - Licence: mit (https://huggingface.co/microsoft/Phi-4-reasoning/resolve/main/LICENSE) - Open weights: yes - Release date: 2025-04-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Phi 4 Reasoning is published by Microsoft as open weights with 14.7B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-4-reasoning, https://huggingface.co/api/models/microsoft/Phi-4-reasoning?blobs=true, https://huggingface.co/microsoft/Phi-4-reasoning/resolve/main/LICENSE ## Phi Mini MoE Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-mini-moe-instruct/ - Context window: 0 - Parameters: 7.6B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-06-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Phi Mini MoE Instruct is published by Microsoft as open weights with 7.6B parameters under MIT. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-mini-MoE-instruct, https://huggingface.co/api/models/microsoft/Phi-mini-MoE-instruct?blobs=true, https://opensource.org/license/mit ## Phi Tiny MoE Instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-tiny-moe-instruct/ - Context window: 0 - Parameters: 3.8B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-06-23 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Phi Tiny MoE Instruct is published by Microsoft as open weights with 3.8B parameters under MIT. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/microsoft/Phi-tiny-MoE-instruct, https://huggingface.co/api/models/microsoft/Phi-tiny-MoE-instruct?blobs=true, https://opensource.org/license/mit ## Phi-4 - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-4/ - Context window: 16384 - Parameters: 14.66B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2024-12-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Phi-4 is Microsoft’s 14B text model, released under the MIT licence with a 16K-token context window and a declared English-only language list. The model card reports 84.8 on MMLU, 56.1 on GPQA and 82.6 on HumanEval with synthetic-data-heavy training. It remains a compact option for reasoning and maths at low hardware cost, but the context window is small by current standards. Sources: https://huggingface.co/microsoft/phi-4, https://arxiv.org/abs/2412.08905 ## Phi-4-multimodal-instruct - Provider: Microsoft - Page: https://llmeu.com/en/models/phi-4-multimodal-instruct/ - Context window: 131072 - Parameters: 5.57B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2025-02-24 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Phi-4-multimodal-instruct is a 5.6B model that accepts text, images and speech in one checkpoint, with a 128K-token context and the MIT licence. The model card reports 55.1 on MMMU, 93.2 on DocVQA and 62.4 on MathVista, and lists the supported languages separately for each input modality. It is a compact option for document and speech pipelines that would otherwise need two models. Sources: https://huggingface.co/microsoft/Phi-4-multimodal-instruct, https://arxiv.org/abs/2503.01743 ## Pixtral 12B - Provider: Mistral AI - Page: https://llmeu.com/en/models/pixtral-12b/ - Context window: 131072 - Parameters: 12B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-09-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Pixtral 12B is Mistral’s first vision-language model, released as open weights under Apache-2.0. Its configuration file declares a maximum position embedding of 131,072 tokens, and the card reports 52.5 on MMMU (chain of thought) and 90.7 on DocVQA. It has been superseded by later Mistral multimodal and Ministral releases but remains downloadable. Sources: https://huggingface.co/mistralai/Pixtral-12B-2409 ## Pythia 1.4B - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-1-4b/ - Context window: 0 - Parameters: 1.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Pythia 1.4B is a small open-weight language model from EleutherAI: 1.5B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-1.4b, https://huggingface.co/api/models/EleutherAI/pythia-1.4b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 12B - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-12b/ - Context window: 0 - Parameters: 12B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Pythia 12B is a large open-weight language model from EleutherAI: 12B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-12b, https://huggingface.co/api/models/EleutherAI/pythia-12b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 160M - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-160m/ - Context window: 0 - Parameters: 0.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-08 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Pythia 160M is a small open-weight language model from EleutherAI: 0.2B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-160m, https://huggingface.co/api/models/EleutherAI/pythia-160m?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 1B - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-1b/ - Context window: 0 - Parameters: 1.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-03-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Pythia 1B is a small open-weight language model from EleutherAI: 1.1B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in March 2023 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-1b, https://huggingface.co/api/models/EleutherAI/pythia-1b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 2.8B - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-2-8b/ - Context window: 0 - Parameters: 2.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Pythia 2.8B is a mid-size open-weight language model from EleutherAI: 2.9B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-2.8b, https://huggingface.co/api/models/EleutherAI/pythia-2.8b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 410M - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-410m/ - Context window: 0 - Parameters: 0.5B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Pythia 410M is a small open-weight language model from EleutherAI: 0.5B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-410m, https://huggingface.co/api/models/EleutherAI/pythia-410m?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 6.9B - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-6-9b/ - Context window: 0 - Parameters: 7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-14 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Pythia 6.9B is a mid-size open-weight language model from EleutherAI: 7B parameters under Apache-2.0. The repository was created in February 2023 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-6.9b, https://huggingface.co/api/models/EleutherAI/pythia-6.9b?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Pythia 70M - Provider: EleutherAI - Page: https://llmeu.com/en/models/pythia-70m/ - Context window: 0 - Parameters: 0.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-02-13 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Pythia 70M is a small open-weight language model from EleutherAI: 0.1B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in February 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/EleutherAI/pythia-70m, https://huggingface.co/api/models/EleutherAI/pythia-70m?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qianfan OCR - Provider: Baidu - Page: https://llmeu.com/en/models/qianfan-ocr/ - Context window: 0 - Parameters: 4.7B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-03-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qianfan OCR is an open-weight vision-language model from Baidu with 4.7B parameters, published under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in March 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/baidu/Qianfan-OCR, https://huggingface.co/api/models/baidu/Qianfan-OCR?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen2.5 0.5B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-0-5b-instruct/ - Context window: 0 - Parameters: 0.5B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 0.5B Instruct is an open-weight language model from Alibaba Qwen with 0.5B parameters, published under Apache-2.0. The repository was created in September 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-0.5B-Instruct?blobs=true, https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE ## Qwen2.5 1.5B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-1-5b-instruct/ - Context window: 0 - Parameters: 1.5B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-09-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen2.5 1.5B Instruct is an open-weight language model from Alibaba Qwen with 1.5B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-1.5B-Instruct?blobs=true, https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE ## Qwen2.5 3B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-3b-instruct/ - Context window: 0 - Parameters: 3.1B - Licence: other (https://huggingface.co/Qwen/Qwen2.5-3B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-09-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 3B Instruct is an open-weight language model from Alibaba Qwen with 3.1B parameters, published under qwen-research (custom, tagged "other"). It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-3B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-3B-Instruct?blobs=true, https://huggingface.co/Qwen/Qwen2.5-3B-Instruct/blob/main/LICENSE ## Qwen2.5 72B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-72b-instruct/ - Context window: 0 - Parameters: unverified - Licence: other (https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 72B Instruct is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under other. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-72B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-72B-Instruct?blobs=true, https://huggingface.co/Qwen/Qwen2.5-72B-Instruct/blob/main/LICENSE ## Qwen2.5 7B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-7b-instruct/ - Context window: 0 - Parameters: 7.6B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/blob/main/LICENSE) - Open weights: yes - Release date: 2024-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 7B Instruct is an open-weight language model from Alibaba Qwen with 7.6B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-7B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-7B-Instruct?blobs=true, https://huggingface.co/Qwen/Qwen2.5-7B-Instruct/blob/main/LICENSE ## Qwen2.5 Coder 14B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-coder-14b-instruct/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-11-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 Coder 14B Instruct is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-Coder-14B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-Coder-14B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen2.5 Coder 32B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-coder-32b-instruct/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-11-06 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 Coder 32B Instruct is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-Coder-32B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen2.5 Coder 7B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-coder-7b-instruct/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-09-17 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen2.5 Coder 7B Instruct is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-Coder-7B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen2.5 VL 7B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen2-5-vl-7b-instruct/ - Context window: 0 - Parameters: 8.3B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-01-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Qwen2.5 VL 7B Instruct is a mid-size open-weight vision-language model from Alibaba Qwen: 8.3B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in January 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct, https://huggingface.co/api/models/Qwen/Qwen2.5-VL-7B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 0.6B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-0-6b/ - Context window: 0 - Parameters: 0.8B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-0.6B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen3 0.6B is an open-weight language model from Alibaba Qwen with 0.8B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-0.6B, https://huggingface.co/api/models/Qwen/Qwen3-0.6B?blobs=true, https://huggingface.co/Qwen/Qwen3-0.6B/blob/main/LICENSE ## Qwen3 1.7B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-1-7b/ - Context window: 0 - Parameters: 2B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 1.7B is an open-weight language model from Alibaba Qwen with 2B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-1.7B, https://huggingface.co/api/models/Qwen/Qwen3-1.7B?blobs=true, https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE ## Qwen3 14B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-14b/ - Context window: 0 - Parameters: 14.8B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-14B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen3 14B is an open-weight language model from Alibaba Qwen with 14.8B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-14B, https://huggingface.co/api/models/Qwen/Qwen3-14B?blobs=true, https://huggingface.co/Qwen/Qwen3-14B/blob/main/LICENSE ## Qwen3 30B A3B Instruct 2507 - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-30b-a3b-instruct-2507/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-07-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Qwen3 30B A3B Instruct 2507 is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in July 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507, https://huggingface.co/api/models/Qwen/Qwen3-30B-A3B-Instruct-2507?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 32B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-32b/ - Context window: 0 - Parameters: 32.8B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-32B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen3 32B is an open-weight language model from Alibaba Qwen with 32.8B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-32B, https://huggingface.co/api/models/Qwen/Qwen3-32B?blobs=true, https://huggingface.co/Qwen/Qwen3-32B/blob/main/LICENSE ## Qwen3 4B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-4b/ - Context window: 0 - Parameters: 4B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-4B/blob/main/LICENSE) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3 4B is an open-weight language model from Alibaba Qwen with 4B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in April 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-4B, https://huggingface.co/api/models/Qwen/Qwen3-4B?blobs=true, https://huggingface.co/Qwen/Qwen3-4B/blob/main/LICENSE ## Qwen3 4B Instruct 2507 - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-4b-instruct-2507/ - Context window: 0 - Parameters: 4B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/blob/main/LICENSE) - Open weights: yes - Release date: 2025-08-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Qwen3 4B Instruct 2507 is an open-weight language model from Alibaba Qwen with 4B parameters, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in August 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507, https://huggingface.co/api/models/Qwen/Qwen3-4B-Instruct-2507?blobs=true, https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507/blob/main/LICENSE ## Qwen3 Coder 480B A35B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-coder-480b-a35b-instruct/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-07-22 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen3 Coder 480B A35B Instruct is a unstated size open-weight language model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in July 2025 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct, https://huggingface.co/api/models/Qwen/Qwen3-Coder-480B-A35B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Coder Next - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-coder-next/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 Coder Next is a unstated size open-weight language model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Coder-Next, https://huggingface.co/api/models/Qwen/Qwen3-Coder-Next?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Embedding 0.6B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-embedding-0-6b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It turns text into vectors for retrieval instead of generating answers. Qwen3 Embedding 0.6B is a unstated size open-weight embedding model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Embedding-0.6B, https://huggingface.co/api/models/Qwen/Qwen3-Embedding-0.6B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Embedding 4B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-embedding-4b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3 Embedding 4B is a unstated size open-weight embedding model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Embedding-4B, https://huggingface.co/api/models/Qwen/Qwen3-Embedding-4B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Embedding 8B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-embedding-8b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3 Embedding 8B is a unstated size open-weight embedding model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Embedding-8B, https://huggingface.co/api/models/Qwen/Qwen3-Embedding-8B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Next 80B A3B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-next-80b-a3b-instruct/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Qwen3 Next 80B A3B Instruct is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in September 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct, https://huggingface.co/api/models/Qwen/Qwen3-Next-80B-A3B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Next 80B A3B Thinking - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-next-80b-a3b-thinking/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-09-09 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 Next 80B A3B Thinking is an open-weight language model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. The repository was created in September 2025 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Thinking, https://huggingface.co/api/models/Qwen/Qwen3-Next-80B-A3B-Thinking?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Reranker 0.6B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-reranker-0-6b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-05-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It turns text into vectors for retrieval instead of generating answers. Qwen3 Reranker 0.6B is an open-weight embedding model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Reranker-0.6B, https://huggingface.co/api/models/Qwen/Qwen3-Reranker-0.6B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Reranker 4B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-reranker-4b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-06-03 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3 Reranker 4B is an open-weight embedding model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in June 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Reranker-4B, https://huggingface.co/api/models/Qwen/Qwen3-Reranker-4B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 Reranker 8B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-reranker-8b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-05-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3 Reranker 8B is an open-weight embedding model from Alibaba Qwen with an unpublished parameter count, published under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in May 2025 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-Reranker-8B, https://huggingface.co/api/models/Qwen/Qwen3-Reranker-8B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 TTS 12Hz 1.7B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-tts-12hz-1-7b-base/ - Context window: 0 - Parameters: 1.9B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Qwen3 TTS 12Hz 1.7B is an open-weight language model from Alibaba Qwen with 1.9B parameters, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in January 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-TTS-12Hz-1.7B-Base, https://huggingface.co/api/models/Qwen/Qwen3-TTS-12Hz-1.7B-Base?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 VL 4B Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-4b-instruct/ - Context window: 0 - Parameters: 4.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-10-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 VL 4B Instruct is a mid-size open-weight vision-language model from Alibaba Qwen: 4.4B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct, https://huggingface.co/api/models/Qwen/Qwen3-VL-4B-Instruct?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 VL Embedding 2B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-embedding-2b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 VL Embedding 2B is published by Alibaba Qwen as open weights with an unpublished parameter count under Apache-2.0. The repository was created in January 2026 and is mirrored on Hugging Face. It turns text into vectors for retrieval instead of generating answers. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-VL-Embedding-2B, https://huggingface.co/api/models/Qwen/Qwen3-VL-Embedding-2B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 VL Embedding 8B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-embedding-8b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Qwen3 VL Embedding 8B is published by Alibaba Qwen as open weights with an unpublished parameter count under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in January 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-VL-Embedding-8B, https://huggingface.co/api/models/Qwen/Qwen3-VL-Embedding-8B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 VL Reranker 2B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-reranker-2b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3 VL Reranker 2B is a unstated size open-weight embedding model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. The repository was created in January 2026 and is mirrored on Hugging Face. It turns text into vectors for retrieval instead of generating answers. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-VL-Reranker-2B, https://huggingface.co/api/models/Qwen/Qwen3-VL-Reranker-2B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3 VL Reranker 8B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-reranker-8b/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-01-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Qwen3 VL Reranker 8B is a unstated size open-weight embedding model from Alibaba Qwen: an unpublished parameter count under Apache-2.0. It turns text into vectors for retrieval instead of generating answers. The published configuration records no context window, so this card carries none. The repository was created in January 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3-VL-Reranker-8B, https://huggingface.co/api/models/Qwen/Qwen3-VL-Reranker-8B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Qwen3-8B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-8b/ - Context window: 32768 - Parameters: 8.19B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-04-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3-8B is a dense text model that supports both a thinking and a non-thinking mode in one checkpoint. It handles 32,768 tokens natively, and the model card states that YaRN scaling extends it to 131,072 tokens. It is published under Apache-2.0 and remains one of the most downloaded open text models in its size class. Sources: https://huggingface.co/Qwen/Qwen3-8B ## Qwen3-Coder-30B-A3B-Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-coder-30b-a3b-instruct/ - Context window: 262144 - Parameters: 30.53B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-07-31 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3-Coder-30B-A3B-Instruct is a mixture-of-experts coding model with 30.5B total parameters and roughly 3B activated per token, which keeps agentic coding loops cheap to run. The card states a native context of 262,144 tokens and recommends reducing it on machines with limited memory. It is published under Apache-2.0. Sources: https://huggingface.co/Qwen/Qwen3-Coder-30B-A3B-Instruct ## Qwen3-VL-8B-Instruct - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-vl-8b-instruct/ - Context window: 262144 - Parameters: 8.77B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-10-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3-VL-8B-Instruct is the small vision-language model of the Qwen3-VL generation, with a native 256k-token context that the vendor says can be extended to 1M. It handles images and hours-long video, and is published under Apache-2.0. At 8.8B parameters it is one of the smallest current open models that combine document understanding with long context. Sources: https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct ## Qwen3.5 2B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-5-2b/ - Context window: 0 - Parameters: 2.3B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3.5-2B/blob/main/LICENSE) - Open weights: yes - Release date: 2026-02-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It accepts text and images and answers in text. Qwen3.5 2B is published by Alibaba Qwen as open weights with 2.3B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in February 2026 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3.5-2B, https://huggingface.co/api/models/Qwen/Qwen3.5-2B?blobs=true, https://huggingface.co/Qwen/Qwen3.5-2B/blob/main/LICENSE ## Qwen3.5 4B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-5-4b/ - Context window: 0 - Parameters: 4.7B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3.5-4B/blob/main/LICENSE) - Open weights: yes - Release date: 2026-02-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3.5 4B is published by Alibaba Qwen as open weights with 4.7B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3.5-4B, https://huggingface.co/api/models/Qwen/Qwen3.5-4B?blobs=true, https://huggingface.co/Qwen/Qwen3.5-4B/blob/main/LICENSE ## Qwen3.5 9B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-5-9b/ - Context window: 0 - Parameters: 9.7B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3.5-9B/blob/main/LICENSE) - Open weights: yes - Release date: 2026-02-27 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3.5 9B is published by Alibaba Qwen as open weights with 9.7B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in February 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3.5-9B, https://huggingface.co/api/models/Qwen/Qwen3.5-9B?blobs=true, https://huggingface.co/Qwen/Qwen3.5-9B/blob/main/LICENSE ## Qwen3.6 35B A3B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-6-35b-a3b/ - Context window: 0 - Parameters: 36B - Licence: apache-2.0 (https://huggingface.co/Qwen/Qwen3.6-35B-A3B/blob/main/LICENSE) - Open weights: yes - Release date: 2026-04-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Qwen3.6 35B A3B is a large open-weight vision-language model from Alibaba Qwen: 36B parameters under Apache-2.0. It accepts text and images and answers in text. The published configuration records no context window, so this card carries none. The repository was created in April 2026 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/Qwen/Qwen3.6-35B-A3B, https://huggingface.co/api/models/Qwen/Qwen3.6-35B-A3B?blobs=true, https://huggingface.co/Qwen/Qwen3.6-35B-A3B/blob/main/LICENSE ## Qwen3.8-27B - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/qwen3-8-27b/ - Context window: 262144 - Parameters: 27.78B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-08-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Qwen3.8-27B is a dense vision-language model with a 262,144-token context that the vendor says can be extended to one million tokens. Thinking mode is enabled by default and can be disabled per request, with reasoning depth controlled by a reasoning_effort parameter. It is published under Apache-2.0 and is the smallest model of the Qwen3.8 generation. Sources: https://huggingface.co/Qwen/Qwen3.8-27B, https://www.qwencloud.com/models/qwen3.8-27b ## Stablelm 2 1 6B - Provider: Stability AI - Page: https://llmeu.com/en/models/stablelm-2-1-6b/ - Context window: 0 - Parameters: 1.6B - Licence: other (https://huggingface.co/stabilityai/stablelm-2-1_6b/blob/main/LICENSE) - Open weights: yes - Release date: 2024-01-18 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Stablelm 2 1 6B is a small open-weight language model from Stability AI: 1.6B parameters under other. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 7 languages, 7 of them among the languages the European Union publishes in. Sources: https://huggingface.co/stabilityai/stablelm-2-1_6b, https://huggingface.co/api/models/stabilityai/stablelm-2-1_6b?blobs=true, https://huggingface.co/stabilityai/stablelm-2-1_6b/blob/main/LICENSE ## Stablelm 3B 4e1t - Provider: Stability AI - Page: https://llmeu.com/en/models/stablelm-3b-4e1t/ - Context window: 0 - Parameters: 2.8B - Licence: cc-by-sa-4.0 (https://creativecommons.org/licenses/by-sa/4.0/) - Open weights: yes - Release date: 2023-09-29 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Stablelm 3B 4e1t is a mid-size open-weight language model from Stability AI: 2.8B parameters under cc-by-sa-4.0. The repository was created in September 2023 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/stabilityai/stablelm-3b-4e1t, https://huggingface.co/api/models/stabilityai/stablelm-3b-4e1t?blobs=true, https://creativecommons.org/licenses/by-sa/4.0/ ## Stablelm Zephyr 3B - Provider: Stability AI - Page: https://llmeu.com/en/models/stablelm-zephyr-3b/ - Context window: 0 - Parameters: 2.8B - Licence: other (https://huggingface.co/stabilityai/stablelm-zephyr-3b/blob/main/LICENSE) - Open weights: yes - Release date: 2023-11-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Stablelm Zephyr 3B is published by Stability AI as open weights with 2.8B parameters under other. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in November 2023 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/stabilityai/stablelm-zephyr-3b, https://huggingface.co/api/models/stabilityai/stablelm-zephyr-3b?blobs=true, https://huggingface.co/stabilityai/stablelm-zephyr-3b/blob/main/LICENSE ## Teuken 7B Instruct v0.6 - Provider: OpenGPT-X (Teuken) - Page: https://llmeu.com/en/models/teuken-7b-instruct-v0-6/ - Context window: 4096 - Parameters: 7.45B - Licence: cc-by-nc-4.0 (https://creativecommons.org/licenses/by-nc/4.0/) - Open weights: yes - Release date: 2025-07-28 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Teuken 7B Instruct v0.6 is a German publicly funded model trained on all 24 official EU languages, developed by Fraunhofer, Forschungszentrum Jülich, TU Dresden and DFKI. It is published under CC-BY-NC-4.0, so the weights are downloadable but may only be used for private, non-commercial, research and educational purposes. On the project’s own EU21 evaluation it averages 0.57, ahead of Llama 3.1 8B Instruct at 0.563 and Mistral 7B Instruct v0.3 at 0.527 on the same test set. Sources: https://huggingface.co/openGPT-X/Teuken-7B-instruct-v0.6, https://arxiv.org/abs/2410.08928 ## Unlimited OCR - Provider: Baidu - Page: https://llmeu.com/en/models/unlimited-ocr/ - Context window: 0 - Parameters: 3.3B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2026-06-19 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Unlimited OCR is a mid-size open-weight vision-language model from Baidu: 3.3B parameters under MIT. The repository was created in June 2026 and is mirrored on Hugging Face. It accepts text and images and answers in text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 1 language. Sources: https://huggingface.co/baidu/Unlimited-OCR, https://huggingface.co/api/models/baidu/Unlimited-OCR?blobs=true, https://opensource.org/license/mit ## Voxtral Small 24B - Provider: Mistral AI - Page: https://llmeu.com/en/models/voxtral-small-24b/ - Context window: 32768 - Parameters: 24.26B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-07-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Voxtral Small (voxtral-small-24b) is Mistral’s open-weights audio model under Apache-2.0. It transcribes speech and answers questions about it, with 24.3B parameters and a 32k-token context the vendor maps to about 30 minutes of audio for transcription; it also takes text, detects the source language itself, and covers eight named languages including German. Sources: https://huggingface.co/mistralai/Voxtral-Small-24B-2507, https://arxiv.org/abs/2507.13264 ## WeVisDoc 2B - Provider: Tencent - Page: https://llmeu.com/en/models/wevisdoc-2b/ - Context window: 0 - Parameters: 2.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2026-09-16 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report WeVisDoc 2B is a mid-size open-weight language model from Tencent: 2.4B parameters under Apache-2.0. The repository was created in September 2026 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 2 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/tencent/WeVisDoc-2B, https://huggingface.co/api/models/tencent/WeVisDoc-2B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Base - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-base/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-09-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes audio in and writes text out. Whisper Base is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-base, https://huggingface.co/api/models/openai/whisper-base?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Large - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-large/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-09-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes audio in and writes text out. Whisper Large is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-large, https://huggingface.co/api/models/openai/whisper-large?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Large V2 - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-large-v2/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-12-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It is a base checkpoint without instruction tuning. Whisper Large V2 is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. It takes audio in and writes text out. The published configuration records no context window, so this card carries none. The repository was created in December 2022 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-large-v2, https://huggingface.co/api/models/openai/whisper-large-v2?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Medium - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-medium/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-09-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes audio in and writes text out. Whisper Medium is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-medium, https://huggingface.co/api/models/openai/whisper-medium?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Small - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-small/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-09-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes audio in and writes text out. Whisper Small is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in September 2022 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-small, https://huggingface.co/api/models/openai/whisper-small?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper Tiny - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-tiny/ - Context window: 0 - Parameters: unverified - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2022-09-26 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Whisper Tiny is an open-weight speech model from OpenAI with an unpublished parameter count, published under Apache-2.0. The repository was created in September 2022 and is mirrored on Hugging Face. It takes audio in and writes text out. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher lists 101 languages, 23 of them among the languages the European Union publishes in. Sources: https://huggingface.co/openai/whisper-tiny, https://huggingface.co/api/models/openai/whisper-tiny?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Whisper large-v3 - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-large-v3/ - Context window: unverified - Parameters: 1.54B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-11-07 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Whisper large-v3 is OpenAI’s multilingual speech-recognition and speech-translation model with 1.54B parameters under the Apache-2.0 licence. It has a fixed 30-second receptive field and processes longer audio with a sliding window, and it uses 128 mel frequency bins rather than the 80 of earlier versions. The model card warns that it was evaluated on roughly ten languages and should be evaluated in-domain before deployment. Sources: https://huggingface.co/openai/whisper-large-v3, https://arxiv.org/abs/2212.04356 ## Whisper large-v3-turbo - Provider: OpenAI - Page: https://llmeu.com/en/models/whisper-large-v3-turbo/ - Context window: unverified - Parameters: 0.81B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2024-10-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Whisper large-v3-turbo is a distilled version of Whisper large-v3 with 809M parameters, released under the MIT licence. It keeps the same 30-second receptive field and the multilingual language coverage of large-v3 but decodes faster, which is the reason to prefer it where latency matters. The model card is the shared Whisper card and publishes no benchmark table specific to this variant. Sources: https://huggingface.co/openai/whisper-large-v3-turbo ## Wildguard - Provider: Allen Institute for AI (Ai2) - Page: https://llmeu.com/en/models/wildguard/ - Context window: 0 - Parameters: 7.2B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-06-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Wildguard is published by Allen Institute for AI (Ai2) as open weights with 7.2B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in June 2024 and is mirrored on Hugging Face. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher lists 3 languages, 1 of them among the languages the European Union publishes in. Sources: https://huggingface.co/allenai/wildguard, https://huggingface.co/api/models/allenai/wildguard?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 1.5 34B Chat - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-1-5-34b-chat/ - Context window: 0 - Parameters: 34.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-05-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Yi 1.5 34B Chat is a large open-weight language model from 01.AI: 34.4B parameters under Apache-2.0. The repository was created in May 2024 and is mirrored on Hugging Face. It takes text and returns text. The checkpoint is instruction-tuned and ships a chat template. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-1.5-34B-Chat, https://huggingface.co/api/models/01-ai/Yi-1.5-34B-Chat?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 1.5 6B Chat - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-1-5-6b-chat/ - Context window: 0 - Parameters: 6.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-05-11 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Yi 1.5 6B Chat is a mid-size open-weight language model from 01.AI: 6.1B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The checkpoint is instruction-tuned and ships a chat template. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-1.5-6B-Chat, https://huggingface.co/api/models/01-ai/Yi-1.5-6B-Chat?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 1.5 9B Chat - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-1-5-9b-chat/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-05-10 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Yi 1.5 9B Chat is a mid-size open-weight language model from 01.AI: 8.8B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in May 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-1.5-9B-Chat, https://huggingface.co/api/models/01-ai/Yi-1.5-9B-Chat?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 34B - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-34b/ - Context window: 0 - Parameters: 34.4B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-11-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Yi 34B is a large open-weight language model from 01.AI: 34.4B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-34B, https://huggingface.co/api/models/01-ai/Yi-34B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 6B - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-6b/ - Context window: 0 - Parameters: 6.1B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2023-11-01 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Yi 6B is a mid-size open-weight language model from 01.AI: 6.1B parameters under Apache-2.0. The repository was created in November 2023 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-6B, https://huggingface.co/api/models/01-ai/Yi-6B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi 9B 200K - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-9b-200k/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-03-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Yi 9B 200K is a mid-size open-weight language model from 01.AI: 8.8B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-9B-200K, https://huggingface.co/api/models/01-ai/Yi-9B-200K?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi-Coder 9B - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-coder-9b/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-08-15 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report It takes text and returns text. Yi-Coder 9B is a mid-size open-weight language model from 01.AI: 8.8B parameters under Apache-2.0. The published configuration records no context window, so this card carries none. The repository was created in August 2024 and is mirrored on Hugging Face. It is a base checkpoint without instruction tuning. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-Coder-9B, https://huggingface.co/api/models/01-ai/Yi-Coder-9B?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Yi-Coder 9B Chat - Provider: 01.AI - Page: https://llmeu.com/en/models/yi-coder-9b-chat/ - Context window: 0 - Parameters: 8.8B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-08-21 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report The checkpoint is instruction-tuned and ships a chat template. Yi-Coder 9B Chat is a mid-size open-weight language model from 01.AI: 8.8B parameters under Apache-2.0. It takes text and returns text. The published configuration records no context window, so this card carries none. The repository was created in August 2024 and is mirrored on Hugging Face. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/01-ai/Yi-Coder-9B-Chat, https://huggingface.co/api/models/01-ai/Yi-Coder-9B-Chat?blobs=true, https://www.apache.org/licenses/LICENSE-2.0 ## Youtu LLM 2B - Provider: Tencent - Page: https://llmeu.com/en/models/youtu-llm-2b/ - Context window: 0 - Parameters: 2B - Licence: other (https://huggingface.co/tencent/Youtu-LLM-2B/LICENSE.txt) - Open weights: yes - Release date: 2025-12-31 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Youtu LLM 2B is published by Tencent as open weights with 2B parameters under youtu-llm (custom, tagged "other"). The repository was created in December 2025 and is mirrored on Hugging Face. It takes text and returns text. It is a base checkpoint without instruction tuning. The published configuration records no context window, so this card carries none. The card is catalogue-only: LLM EU does not run this model. The publisher states no language coverage on the model page. Sources: https://huggingface.co/tencent/Youtu-LLM-2B, https://huggingface.co/api/models/tencent/Youtu-LLM-2B?blobs=true, https://huggingface.co/tencent/Youtu-LLM-2B/LICENSE.txt ## gpt-oss-120b - Provider: OpenAI - Page: https://llmeu.com/en/models/gpt-oss-120b/ - Context window: 131072 - Parameters: 116.83B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-08-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report gpt-oss-120b is OpenAI’s open-weights reasoning model: a mixture of experts with 117B total and 5.1B activated per token, under Apache-2.0. Its 131,072-token context window and expert weights post-trained in MXFP4 let it fit on a single 80GB GPU, and it needs OpenAI’s harmony response format with three reasoning-effort levels. Sources: https://huggingface.co/openai/gpt-oss-120b, https://arxiv.org/abs/2508.10925 ## gpt-oss-20b - Provider: OpenAI - Page: https://llmeu.com/en/models/gpt-oss-20b/ - Context window: 131072 - Parameters: 20.91B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2025-08-05 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report gpt-oss-20b is the smaller of OpenAI’s two open-weight reasoning models: 21B total parameters, 3.6B active, Apache-2.0 licensed and small enough to run in about 16GB of memory. Its configuration declares the same 131,072-token context window as the 120B model. It targets local and latency-sensitive deployments rather than peak quality. Sources: https://huggingface.co/openai/gpt-oss-20b ## gte-multilingual-reranker-base - Provider: Alibaba Qwen - Page: https://llmeu.com/en/models/gte-multilingual-reranker-base/ - Context window: 8192 - Parameters: 0.31B - Licence: apache-2.0 (https://www.apache.org/licenses/LICENSE-2.0) - Open weights: yes - Release date: 2024-07-20 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report gte-multilingual-reranker-base is a cross-encoder reranker for multilingual retrieval, with 305M parameters and support for inputs up to 8192 tokens. It is published under Apache-2.0 and covers 75 languages in its declared language list. In a retrieval pipeline it belongs after a first-stage vector search, not as a standalone embedding model. Sources: https://huggingface.co/Alibaba-NLP/gte-multilingual-reranker-base, https://arxiv.org/abs/2407.19669 ## inclusionAI: Ling 3.0 Flash Sante (free) - Provider: OpenRouter - Page: https://llmeu.com/en/models/ling-3.0-flash-sante:free/ - Context window: 262144 - Parameters: unverified - Licence: unverified - Open weights: no - Release date: unverified - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report Served through OpenRouter as inclusionai/ling-3.0-flash-sante:free. Runs outside the EU; the residency class on its endpoint says so. Sources: https://openrouter.ai/models ## multilingual-e5-large - Provider: Microsoft - Page: https://llmeu.com/en/models/multilingual-e5-large/ - Context window: 512 - Parameters: 0.56B - Licence: mit (https://opensource.org/license/mit) - Open weights: yes - Release date: 2023-06-30 - Hosting: catalogue only — not hosted on LLM EU yet; no live endpoint, no region and no retention to report multilingual-e5-large is an XLM-RoBERTa-based text embedding model with 560M parameters, published under the MIT licence and covering around 100 languages. The model card uses a maximum sequence length of 512 tokens, which is the effective limit for both indexing and querying. It has no instruction prefix for queries, so documents and queries must be embedded the same way. Sources: https://huggingface.co/intfloat/multilingual-e5-large, https://arxiv.org/abs/2402.05672 ## Providers - **Aleph Alpha** (DE) — European provider — https://llmeu.com/en/providers/aleph-alpha/ - **EuroLLM (utter-project)** (PT) — European provider — https://llmeu.com/en/providers/eurollm/ - **LLM EU** (DE) — European provider — https://llmeu.com/en/providers/llmeu/ - **Mistral AI** (FR) — European provider — https://llmeu.com/en/providers/mistral/ - **OpenEuroLLM** (CZ) — European provider — https://llmeu.com/en/providers/openeurollm/ - **OpenGPT-X (Teuken)** (DE) — European provider — https://llmeu.com/en/providers/opengptx/ - **01.AI** (CN) — non-European provider — https://llmeu.com/en/providers/01-ai/ - **Alibaba Qwen** (CN) — non-European provider — https://llmeu.com/en/providers/alibaba-qwen/ - **Allen Institute for AI (Ai2)** (US) — non-European provider — https://llmeu.com/en/providers/allenai/ - **Baidu** (CN) — non-European provider — https://llmeu.com/en/providers/baidu/ - **BigScience (BLOOM)** (US) — non-European provider — https://llmeu.com/en/providers/bigscience/ - **Cohere** (CA) — non-European provider — https://llmeu.com/en/providers/cohere/ - **DeepSeek** (CN) — non-European provider — https://llmeu.com/en/providers/deepseek/ - **EleutherAI** (US) — non-European provider — https://llmeu.com/en/providers/eleutherai/ - **Google DeepMind** (US) — non-European provider — https://llmeu.com/en/providers/google/ - **IBM** (US) — non-European provider — https://llmeu.com/en/providers/ibm/ - **Meta** (US) — non-European provider — https://llmeu.com/en/providers/meta/ - **Microsoft** (US) — non-European provider — https://llmeu.com/en/providers/microsoft/ - **MiniMax** (CN) — non-European provider — https://llmeu.com/en/providers/minimax/ - **Moonshot AI** (CN) — non-European provider — https://llmeu.com/en/providers/moonshotai/ - **NVIDIA** (US) — non-European provider — https://llmeu.com/en/providers/nvidia/ - **OpenAI** (US) — non-European provider — https://llmeu.com/en/providers/openai/ - **OpenRouter** (US) — non-European provider — https://llmeu.com/en/providers/openrouter/ - **Salesforce** (US) — non-European provider — https://llmeu.com/en/providers/salesforce/ - **Shanghai AI Laboratory** (CN) — non-European provider — https://llmeu.com/en/providers/internlm/ - **Stability AI** (GB) — non-European provider — https://llmeu.com/en/providers/stabilityai/ - **Technology Innovation Institute (TII)** (AE) — non-European provider — https://llmeu.com/en/providers/tiiuae/ - **Tencent** (CN) — non-European provider — https://llmeu.com/en/providers/tencent/ - **Z.ai (Zhipu AI)** (CN) — non-European provider — https://llmeu.com/en/providers/zai-org/ ## Regions - **Berlin, Germany — LLM EU operated** — eu_owned_eu_operated — https://llmeu.com/en/hosting/de-ber/ - **Frankfurt, Germany — LLM EU operated** — eu_owned_eu_operated — https://llmeu.com/en/hosting/de-fra/ - **Paris, France — EU operator** — eu_owned_eu_operated — https://llmeu.com/en/hosting/fr-par/ - **Outside the EU (OpenRouter broker)** — non_eu_third_country — https://llmeu.com/en/hosting/global/