Open-weight · Mistral AI

Mistral Small 4

Mistral's compact 24B model under Apache 2.0 — efficient to run, with a 256K context window, well suited to cost-sensitive self-hosting.

Developer

Mistral AI

Parameters

24B

Context

256K

License

Apache 2.0

About Mistral Small 4

Mistral's compact 24B model under Apache 2.0 — efficient to run, with a 256K context window, well suited to cost-sensitive self-hosting.

Parameters, context, and license as reported June 2026. Confirm the current license on the official Hugging Face page before deployment.

Access hosted models through a simple, metered token API.

Buy or lease Supermicro GPU systems to self-host open-weight models.

Managed compute for training, fine-tuning, and inference.