#pixtral
LightNews LightNews
simonwillison.net
Looks like it's a big LLM release Monday today - so far Qwen 2.5 Turbo (API-only) and a new vision model from Mistral called Pixtral Large (open weights)

qwenlm.github.io/blog/qwen2.5...
mistral.ai/news/pixtral...
Pixtral Large
Pixtral grows up.
mistral.ai
November 18, 2024 at 4:28 PM
merve.bsky.social
Mistral just released a state-of-the-art open-weights 124B vision language model - Pixtral Large

mistral.ai/news/pixtral...
November 18, 2024 at 6:18 PM
reach-vb.hf.co
Great day for M/LLMs, just released Mistral & Pixtral Large - ~123B, 128K context, Multilingual, JSON + Function calling support & open weights! 🔥

Pixtral Large: huggingface.co/mistralai/Pi...

Mistral Large: huggingface.co/mistralai/Mi...
November 18, 2024 at 5:40 PM
simon.fedi.simonwillison.net.ap.brid.gy
Looks like it's a big LLM release Monday today - so far Qwen 2.5 Turbo (API-only) and a new vision model from Mistral called Pixtral Large (open weights)

https://qwenlm.github.io/blog/qwen2.5-turbo/
https://mistral.ai/news/pixtral-large/
November 18, 2024 at 4:31 PM
simonwillison.net
Notes on accessing Pixtral Large via LLM and llm-mistral on my blog: More notes on my blog: simonwillison.net/2024/Nov/18/...
Pixtral Large
New today from Mistral: > Today we announce Pixtral Large, a 124B open-weights multimodal model built on top of Mistral Large 2. Pixtral Large is the second model in our …
simonwillison.net
November 18, 2024 at 4:42 PM
techmeme.com
Mistral releases Small 4, its first model to unify reasoning, multimodal, and coding capabilities of its flagship Magistral, Pixtral, and Devstral models (Mistral AI)

Main Link | Techmeme Permalink
March 17, 2026 at 3:30 AM
haesleinhuepf.bsky.social
Voilà! #gitbob and #biabob were updated to support #pixtral, the vision model from the French mistral.ai #LLM provider. It also comes with a free tier 🇫🇷 🇪🇺 🎉
github.com/haesleinhuep...
github.com/haesleinhuep...
December 1, 2024 at 9:24 AM
reach-vb.hf.co
Massive week for Open Source AI/ ML

Mistral Pixtral & Instruct Large - ~123B, 128K context, multilingual, json + function calling & open weights

Allen AI Tülu 70B & 8B - competive with claude 3.5 haiku, beats all major open models like llama 3.1 70B, qwen 2.5 and nemotron
November 24, 2024 at 8:12 PM
embeddedllm.bsky.social
🔥 Pixtral Large is now supported on vLLM! 🔥

Run Pixtral Large with multiple input images from day 0 using vLLM.

Install vLLM:
pip install -U VLLM

Run Pixtral Large:
vllm serve mistralai/Pixtral-Large-Instruct-2411 --tokenizer_mode mistral --limit_mm_per_prompt 'image=10' --tensor-parallel-size 8
Pixtral Large benchmarks
November 19, 2024 at 12:38 PM
sungkim.bsky.social
Blog: mistral.ai/news/pixtral...

Model Weight: huggingface.co/mistralai/Pi...

Demo on Le Chat: chat.mistral.ai
mistralai/Pixtral-Large-Instruct-2411 · Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
November 18, 2024 at 4:55 PM
dlsq.bsky.social
We have 2 new big updates today at Mistral:
- New Le Chat: With canvas, web search, image understanding and generation & more - and free!
- Pixtral Large, our Frontier 124B open weight multimodal model that powers it.

Try it: chat.mistral.ai
Blog post: mistral.ai/news/mistral...
Two announcement cards from the Mistral AI team, dated November 18, 2024. The first card announces 'Mistral has entered the chat' with a brief description: 'Search, vision, ideation, coding... all yours for free.' The second card announces 'Pixtral Large' with the description: 'Pixtral grows up.' Both cards feature an orange 'Read More' button.
November 18, 2024 at 5:57 PM
techcrunch.com
Mistral releases Pixtral, its first multimodal model
Mistral releases Pixtral, its first multimodal model
French AI startup Mistral has released its first model that can process images as well as text. Called Pixtral 12B, the 12-billion-parameter model is roughly 24GB size. (Parameters roughly correspond to a model’s problem-solving skills, and models with…
tcrn.ch
September 11, 2024 at 12:02 PM
gigazine.net
Mistral AIが124BサイズのマルチモーダルAIモデル「Pixtral Large」リリース、GPT-4oやGemini-1.5 Proなどを上回るパフォーマンス
https://gigazine.net/news/20241119-pixtral-large/
Mistral AIが124BサイズのマルチモーダルAIモデル「Pixtral Large」リリース、GPT-4oやGemini-1.5 Proなどを上回るパフォーマンス
フランスのAIスタートアップであるMistral AIが、既存のMistral Large 2モデルにドキュメント・グラフ・画像の認識能力を加えたマルチモーダルAIモデル「Pixtral Large」をリリースしました。
gigazine.net
November 19, 2024 at 6:00 AM
gplanchat.bsky.social
J'ai créé un serveur #MCP pour m'aider à corriger des bugs d'affichage CSS.

Il utilise Playwright et #Pixtral pour identifier les défauts dans le rendu.

github.com/gplanchat/se...
GitHub - gplanchat/server-playwright: MCP server for analyzing HTML pages using Mistral's Pixtral
MCP server for analyzing HTML pages using Mistral's Pixtral - gplanchat/server-playwright
github.com
December 4, 2025 at 11:05 PM
sungkim.bsky.social
Mistral releases Pixtral Large: Frontier-class 124B multimodal model, powering the new Le Chat.

- Frontier-class multimodal model
- SoTA on MathVista, DocVQA, VQAv2
- Maintains text performance of Mistral Large 2
- 123B decoder, 1B vision encoder
- 128K seq len
November 18, 2024 at 4:55 PM
sungkim.bsky.social
Mistral.ai's Pixtral-12B

The Pixtral-12B-2409 is a Multimodal Model of 12B parameters plus a 400M parameter vision encoder.

Base: huggingface.co/mistralai/Pi...
Instruct: huggingface.co/mistralai/Pi...
October 25, 2024 at 3:29 PM
alexf.bsky.social
Mistral just released Pixtral, a 124B open-weights multimodal model built on top of Mistral Large 2.
mistral.ai/news/pixtral...
Pixtral Large
Pixtral grows up.
mistral.ai
November 20, 2024 at 4:46 PM
simon.fedi.simonwillison.net.ap.brid.gy
Mistral launched Pixtral 12B today, their first multi-modal (text plus images) vision LLM. It’s Apache 2 licensed and a 25GB download. I put together some notes on the release here: https://simonwillison.net/2024/Sep/11/pixtral/
September 11, 2024 at 10:29 PM
zandl.bsky.social
🤖AI - Mistral AI má Pixtral Large a Mistral Large 24.11 a rozšiřuje možnosti: vyhledávání, canvas, zpracování dokumentů, obrázky svého open source modelu.
Mistral AI má Pixtral Large a Mistral Large 24.11 a rozšiřuje možnosti: vyhledávání, canvas, zpracování dokumentů, obrázky
Mistral AI má Pixtral Large a Mistral Large 24.11 a rozšiřuje možnosti: vyhledávání, canvas, zpracování dokumentů, obrázky
buff.ly
November 19, 2024 at 10:48 AM
sungkim.bsky.social
This paper describes Pixtral 12B, a multimodal language model trained to understand both images and text, released with open weights under an Apache 2.0 license.
October 10, 2024 at 3:03 AM
dlsq.bsky.social
Pixtral Large:
- 123B decoder, 1B vision encoder, 128K sequence length
- Frontier multimodal model
- Maintains text performance of Mistral Large 2

HF weights: huggingface.co/mistralai/Pi...
Try it: chat.mistral.ai
Blog post: mistral.ai/news/pixtral...
Comparison table of various AI models across different benchmarks: Mathvista, MMMU, ChartQA, DocVQA, VQAv2, AI2D, and MM MT-Bench. Models are categorized into Open Weights, Closed, and Unreleased. Key models include Pixtral Large, Llama-3.2 90B, Gemini-1.5 Pro, GPT-4o, Claude-3.5 Sonnet, Llama-3.1 505B, and Grok-2. The table shows measured and reported performance scores, highlighting differences in model capabilities across various tasks. Pixtral Large excels in Mathvista, DocVQA, AI2D and MM MT-Bench benchmarks.
November 18, 2024 at 5:57 PM
testingcatalog.com
BREAKING 🚨: Pixtral-Large-Instruct-2411 now appears on huggingface 🔥
Evals 👀👀👀
November 18, 2024 at 4:50 PM
testingcatalog.com
According to the "Limits" page on Le Platforme, Mistral AI is preparing to release 3 new models (supposedly next week?).
- mistral-large-2411 - Potentially a multimodal Large 2.1
- pixtral-large-2411
- mistral-moderation-2411 - a mini model for moderation
November 16, 2024 at 7:48 PM