Introducing Mistral Small 4 (https://mistral.ai/news/mistral-small-4)

Big new release from Mistral today (despite the name) - a new Apache 2 licensed 119B parameter (Mixture-of-Experts, 6B active) model which they describe like this:

Mistral Small 4 is the first Mistral model to unify the capabilities of our flagship models, Magistral for reasoning, Pixtral for multimodal, and Devstral for agentic coding, into a single, versatile model.

It supports reasoning_effort=“none” or reasoning_effort=“high”, with the latter providing “equivalent verbosity to previous Magistral models”.

The new model is 242GB on Hugging Face (https://huggingface.co/mistralai/Mistral-Small-4-119B-2603/tree/main).

I tried it out (https://gist.github.com/simonw/3dec228577559f15f26204a3cc550583) via the Mistral API using llm-mistral (https://github.com/simonw/llm-mistral):

llm install llm-mistral llm mistral refresh llm -m mistral/mistral-small-2603 “Generate an SVG of a pelican riding a bicycle”

I couldn’t find a way to set the reasoning effort in their API documentation (https://docs.mistral.ai/api/endpoint/chat#operation-chat_completion_v1_chat_completions_post), so hopefully that’s a feature which will land soon.

Also from Mistral today and fitting their -stral naming convention is Leanstral (https://mistral.ai/news/leanstral), an open weight model that is specifically tuned to help output the Lean 4 (https://lean-lang.org/) formally verifiable coding language. I haven’t explored Lean at all so I have no way to credibly evaluate this, but it’s interesting to see them target one specific language in this way.

Tags: ai (https://simonwillison.net/tags/ai), generative-ai (https://simonwillison.net/tags/generative-ai), llms (https://simonwillison.net/tags/llms), llm (https://simonwillison.net/tags/llm), mistral (https://simonwillison.net/tags/mistral), pelican-riding-a-bicycle (https://simonwillison.net/tags/pelican-riding-a-bicycle), llm-reasoning (https://simonwillison.net/tags/llm-reasoning), llm-release (https://simonwillison.net/tags/llm-release)