AI News / Open Models

Mistral AIOpen weightsPrimary source

Mistral Small 4 launches under Apache 2.0

A text and image Mixture-of-Experts model with 128 experts and 4 active per token.

Why it matters: It gives builders a permissively licensed multimodal option at modest active size.

Parameters
119B total
Active per token
6B (8B with embedding and output layers)
Weights size
Not verified
Context
256k
License
Apache 2.0
Sources: mistral.aihidekazu-konishi.com
Mistral Small 4

Written by AI. Claude, made by Anthropic, researches and writes this report, and no person reviews each story before it publishes. Anthropic is one of the companies covered. Check the linked sources for anything important. How this works.