Title: Phi Tiny MoE Instruct vs Phi 3.5 Vision Instruct %E2%80%94 LLM EU
Canonical: https://llmeu.com/cs/compare/phi-tiny-moe-instruct-vs-phi-3-5-vision-instruct/
Markdown mirror: https://llmeu.com/cs/compare/phi-tiny-moe-instruct-vs-phi-3-5-vision-instruct.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json

# Phi Tiny MoE Instruct vs Phi 3.5 Vision Instruct

Porovnejte dva a%C5%BE %C4%8Dty%C5%99i modely vedle sebe podle pol%C3%AD, kter%C3%A1 rozhoduj%C3%AD o nasazen%C3%AD: kontext, cena, region, licence, jazyky, hosting a co je ov%C4%9B%C5%99eno.

Tam, kde se dva modely li%C5%A1%C3%AD, tabulka uk%C3%A1%C5%BEe rozd%C3%ADl a skon%C4%8D%C3%AD. LLM EU nezve%C5%99ej%C5%88uje odznak v%C3%ADt%C4%9Bze bez dolo%C5%BEen%C3%A9ho sk%C3%B3re pro danou %C3%BAlohu a chyb%C4%9Bj%C3%ADc%C3%AD %C3%BAdaj z%C5%AFst%C3%A1v%C3%A1 chyb%C4%9Bj%C3%ADc%C3%AD, m%C3%ADsto aby se stal nulou.

## Kdy pou%C5%BE%C3%ADt

### [Phi Tiny MoE Instruct](/cs/models/phi-tiny-moe-instruct/)

#### Siln%C3%A9 str%C3%A1nky

- open weights are published on Hugging Face and can be downloaded without a vendor agreement
- mit licence, so commercial self-hosting is permitted without a separate agreement
- at 3.8B parameters it fits on a single accelerator
- the model page lists en
- instruction-tuned checkpoint with a chat template in the repository

#### Omezen%C3%AD

- catalogue only %E2%80%94 LLM EU does not run this model, and nothing on this card implies an LLM EU endpoint
- no evaluation has been run by LLM EU
- the published configuration records no context window, so the card carries none
- at 3.8B parameters, the weights fit on a single modest accelerator

#### %C3%9Alohy

[Uva%C5%BEov%C3%A1n%C3%AD](/cs/tasks/reasoning/), [N%C3%ADzk%C3%A9 n%C3%A1klady](/cs/tasks/cheap/)

### [Phi 3.5 Vision Instruct](/cs/models/phi-3-5-vision-instruct/)

#### Siln%C3%A9 str%C3%A1nky

- open weights are published on Hugging Face and can be downloaded without a vendor agreement
- mit licence, so commercial self-hosting is permitted without a separate agreement
- at 4.1B parameters it fits on a single accelerator
- instruction-tuned checkpoint with a chat template in the repository

#### Omezen%C3%AD

- catalogue only %E2%80%94 LLM EU does not run this model, and nothing on this card implies an LLM EU endpoint
- no evaluation has been run by LLM EU
- the published configuration records no context window, so the card carries none
- at 4.1B parameters, the weights fit on a single modest accelerator
- vision checkpoints are heavier to serve than their parameter count alone suggests, because the image encoder and pre-processing run alongside the language model

#### %C3%9Alohy

[Zpracov%C3%A1n%C3%AD obrazu](/cs/tasks/vision/), [Shrnut%C3%AD](/cs/tasks/summarization/)

**%C3%9Alohy:** Neov%C4%9B%C5%99eno: poskytovatel toto tvrzen%C3%AD nedokumentoval.

## Porovnat modely

- [Aya 23 8B vs Phi 3.5 Vision Instruct](/cs/compare/aya-23-8b-vs-phi-3-5-vision-instruct/)
- [Aya Expanse 8B vs Phi 3.5 Vision Instruct](/cs/compare/aya-expanse-8b-vs-phi-3-5-vision-instruct/)
- [Blip2 Flan T5 Xl vs Phi 3.5 Vision Instruct](/cs/compare/blip2-flan-t5-xl-vs-phi-3-5-vision-instruct/)
- [Blip2 Opt 2.7B vs Phi Tiny MoE Instruct](/cs/compare/blip2-opt-2-7b-vs-phi-tiny-moe-instruct/)
- [NVIDIA Cosmos 3 Edge vs Phi 3.5 Vision Instruct](/cs/compare/cosmos-3-edge-vs-phi-3-5-vision-instruct/)
- [NVIDIA Cosmos Reason2 2B vs Phi Tiny MoE Instruct](/cs/compare/cosmos-reason2-2b-vs-phi-tiny-moe-instruct/)