Launched Voxtral TTS, a 4B-parameter open-weight text-to-speech model with voice cloning from 3-second audio

Launched Voxtral TTS, its first open-weight text-to-speech model for voice cloning and edge deployment.

Adopted client-first MCP strategy, focusing on MCP consumer rather than provider

Released Mistral Small 4 model via its commercial API platform

Launched Flows, a unified AI creative suite integrating image, video, voice, music, and sound effects generation

Launch of Expressive Mode for ElevenAgents platform, enabling AI voice agents to dynamically adapt tone and emotion

Secured a multi-year partnership with Accenture to deploy its open-source models for enterprise clients.

Deleted Magistral, Pixtral, and Devst models from Hugging Face Hub without announcement