Title: European LLM models: catalogue and hosting %E2%80%94 LLM EU
Canonical: https://llmeu.com/en/models/
Markdown mirror: https://llmeu.com/en/models.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json

# The European model catalogue

Models published by European providers, models hosted in the EU, and models from elsewhere that European teams can run here. Every card states what is documented and what is not.

Run these models yourself %E2%80%94 join the [compute swarm](/en/compute/), a blueprint rather than a GPU provider. The[licence-eligible model index](/en/compute/models/) names the rows you may self-host commercially.

This page is a static copy: the URL you see cannot change what it renders, so there is no filter form here. All 617 models are listed below, and the console offers the same catalogue with filters.

Models found: 617

### [Aya 23 8B](/en/models/aya-23-8b/)

Aya 23 8B is the 8B member of Cohere Labs%E2%80%99 Aya 23 release, covering 23 languages under CC-BY-NC-4.0 with a context length of 8192 tokens. It is the mu%E2%80%A6

Context window: 8.2k %C2%B7 License: cc-by-nc-4.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Aya Expanse 8B](/en/models/aya-expanse-8b/)

Aya Expanse 8B is Cohere Labs%E2%80%99 multilingual instruction model covering 23 languages, released under CC-BY-NC-4.0. Its metadata declares 23 languages, %E2%80%A6

Context window: 8.2k %C2%B7 License: cc-by-nc-4.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Blip2 Flan T5 Xl](/en/models/blip2-flan-t5-xl/)

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 ans%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Blip2 Opt 2.7B](/en/models/blip2-opt-2-7b/)

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%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Bloom](/en/models/bloom/)

It takes text and returns text. Bloom is a very large open-weight language model from BigScience (BLOOM): 176.2B parameters under bigscience-bloom-rai%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: bigscience-bloom-rail-1.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Bloom 1b1](/en/models/bloom-1b1/)

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 i%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: bigscience-bloom-rail-1.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Bloom 3B](/en/models/bloom-3b/)

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%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: bigscience-bloom-rail-1.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Bloom 560M](/en/models/bloom-560m/)

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 %E2%80%A6

Context window: %E2%80%94 %C2%B7 License: bigscience-bloom-rail-1.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Bloomz 560M](/en/models/bloomz-560m/)

The checkpoint is instruction-tuned and ships a chat template. Bloomz 560M is a small open-weight language model from BigScience (BLOOM): 0.6B paramet%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: bigscience-bloom-rail-1.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Codestral 25.08](/en/models/codestral-25-08/)

Codestral is Mistral%E2%80%99s code-completion model, served through the Mistral API rather than published as weights. It supports fill-in-the-middle, tool ca%E2%80%A6

Context window: 131k %C2%B7 License: proprietary %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Command A+ (05-2026)](/en/models/command-a-plus-05-2026/)

Command A+ (command-a-plus-05-2026) is Cohere%E2%80%99s open-weights 219B-parameter model under Apache-2.0, taking text and images and declaring 48 languages.%E2%80%A6

Context window: 131k %C2%B7 License: apache-2.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Command R (08-2024)](/en/models/command-r-08-2024/)

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 %E2%80%A6

Context window: 131k %C2%B7 License: cc-by-nc-4.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [Deepseek Coder 7B Instruct V1.5](/en/models/deepseek-coder-7b-instruct-v1-5/)

Deepseek Coder 7B Instruct V1.5 is an open-weight language model from DeepSeek with 6.9B parameters, published under deepseek (custom, tagged "other")%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: other %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek Coder V2 Lite Instruct](/en/models/deepseek-coder-v2-lite-instruct/)

DeepSeek Coder V2 Lite Instruct is an open-weight language model from DeepSeek with 15.7B parameters, published under deepseek-license (custom, tagged%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: other %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek OCR](/en/models/deepseek-ocr/)

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 i%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek OCR 2](/en/models/deepseek-ocr-2/)

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 J%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: apache-2.0 %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 0528](/en/models/deepseek-r1-0528/)

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 publi%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 0528 Qwen3 8B](/en/models/deepseek-r1-0528-qwen3-8b/)

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 publi%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 Distill Llama 8B](/en/models/deepseek-r1-distill-llama-8b/)

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%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 Distill Qwen 1.5B](/en/models/deepseek-r1-distill-qwen-1-5b/)

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%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 Distill Qwen 14B](/en/models/deepseek-r1-distill-qwen-14b/)

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%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek R1 Distill Qwen 7B](/en/models/deepseek-r1-distill-qwen-7b/)

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. %E2%80%A6

Context window: %E2%80%94 %C2%B7 License: mit %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek V2 Lite](/en/models/deepseek-v2-lite/)

DeepSeek V2 Lite is a large open-weight language model from DeepSeek: 15.7B parameters under deepseek (custom, tagged "other"). The repository was cre%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: other %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

### [DeepSeek V3](/en/models/deepseek-v3/)

DeepSeek V3 is an open-weight language model from DeepSeek with an unpublished parameter count, published under unknown. It takes text and returns tex%E2%80%A6

Context window: %E2%80%94 %C2%B7 License: unknown %C2%B7 Catalogue only

Last reviewed: 14 Oct 2026

These four labels replace the bare word sovereign, which we do not use on its own. A model is called sovereign only with its class and region beside it, because ownership, operation and jurisdiction are three different facts.