Mistral Small 3.1 24B

Good fit for

People evaluating a larger local vision-and-text assistant instead of a small text-only model.

Why consider it

The 24B catalog entry documents image understanding, conversational use and function calling.

Main limitations

Requires Ollama 0.6.5 or later. The model supports up to 128K context, but the published runtime parameters default to 4,096; larger contexts need configuration and memory. A 15GB package is not a 15GB RAM guarantee. Historical speed claims are not published as expected user performance.

Cost and access

No subscription or hosted token fee for local inference within permitted uses. Hardware, electricity and setup are separate. Apache 2.0 is documented for this release.

Getting started

Install a compatible Ollama version, review the model license, then run: ollama run mistral-small3.1:24b. API means the local runtime API, not a hosted endpoint included with the model.