Not established from the available sources.
MISTRAL AI · MODEL RELEASE TRACKER
Voxtral 3
Voxtral 3 is a released family of speech models and a voice-agent SDK from Mistral AI, combining speech-to-text and text-to-speech with a real-time streaming pipeline for listening, reasoning, tool use, and speaking across many languages. It is available via an EU-hosted API or for self-deployment.CURRENT SNAPSHOT2/5 DIMENSIONS WITH DATA
The dimensions that change the decision.
Not established from the available sources.
ModalitiesSpeech: speech-to-text (STT) and text-to-speech (TTS).
Transcription performance claimVoxtral Small Transcribe 3 is described as state-of-the-art in multilingual transcription.
Not established from the available sources.
RELEASE TIMELINE
Published, source-backed release events only.
VERIFIABLE FACTS
Every value stays attached to a source and date.
RELEASE · Launch/announcement dateDEVELOPER CLAIM
Announced/Launched on 2026-10-20.
CAPABILITIES · Core components & capabilitiesDEVELOPER CLAIM
Family includes speech-to-text and text-to-speech models plus a voice agent SDK that performs real-time listening, reasoning, tool calling, and speaking in a single streaming pipeline.
MODALITIES · ModalitiesDEVELOPER CLAIM
Speech: speech-to-text (STT) and text-to-speech (TTS).
CAPABILITIES · Transcription languages supportDEVELOPER CLAIM
Accurate transcription across 50+ languages.
CAPABILITIES · TTS languages, dialects, and emotional adaptationDEVELOPER CLAIM
Text-to-speech supports over 35 languages and 50 dialects with emotional adaptation / expressive speech.
CAPABILITIES · Transcription featuresDEVELOPER CLAIM
Includes diarization (identifying who said what) and context-biasing (using keyphrases to improve accuracy).
CAPABILITIES · Voice agent SDK & tool integrationDEVELOPER CLAIM
Voice agent SDK (reported based on LiveKit) enables deep tool integration and multi-step actions during live conversations in a single streaming pipeline.
BENCHMARKS · Transcription performance claimDEVELOPER CLAIM
Voxtral Small Transcribe 3 is described as state-of-the-art in multilingual transcription.
WHAT CHANGED