Mistral AI Launches Mistral 3 Family of Open Models in Partnership with NVIDIA

Summary

Mistral AI launched its Mistral 3 family of open-weight models on December 2, 2025, in partnership with NVIDIA, marking a significant step in the competition between open-source and proprietary foundation models. The release includes two major components: Mistral Large 3, a mixture-of-experts (MoE) model with 41 billion active parameters and 675 billion total parameters supporting a 256K context window; and the Ministral 3 suite of nine small language models optimized for edge deployment. Mistr