Mistral AI Unveils Voxtral 3 Voice Model and Agent SDK
Current overview (not historical)
Mistral AI has released the Voxtral 3 voice model and a companion Agent SDK, adding real‑time voice transcription, synthesis, and agent capabilities to its product line. The new model supports transcription in over 50 languages and synthesis in 35 languages plus 50 dialects, and includes emotion‑adaptive synthesis. The Agent SDK enables developers to build voice‑driven agents that can call external systems during conversation. Pricing is tiered: the Voxtral Mini Transcribe 3 transcription model costs $0.0035 per minute, while the Voxtral Speech 3 synthesis model is priced at $13 per million characters. Customers can access the services via Mistral EU‑hosted APIs or deploy the models on their own infrastructure.
Written with AI from the reporting · updated 10/22/2026
EventGraph Explorer
One time cutoff for reports, relations and evidence.
New at this step ·Mistral AI launches Voxtral 3 real-time voice agent stack
- Reports in replay
- 2
- First-party
- 1
- Sources
- 2
Reports first seen at this step · 1
- The Decoder · Third-partyMistral AI launches Voxtral 3 real-time voice agent stackOriginal source ↗
Discovery-time replay; the linked graph and evidence share this cutoff. The full overview and activity remain current.
Event developments
1 developments10/21/2026, 12:20:10 AM
Mistral AI launches Voxtral 3, a multilingual voice‑agent SDK with real‑time speech‑to‑text and text‑to‑speech modelsMistral AI
Filtered by first discovery; only representative reports already discovered by this cutoff are shown. Story grouping may have changed later, so this is not a complete historical snapshot.
Relations at this time
Loading time-scoped relations…
Bounded reconstruction, not an exact snapshot: entity names, relationship text and story grouping may include later revisions. Overview and full source archive below remain current.
Story route
A station-by-station route from the first public evidence onward; every station carries its own evidence and uncertainty.
- 1Oct 20, 00:00recorded Oct 21, 00:20revised
Mistral AI 发布 Voxtral 3 语音模型和语音 Agent SDK
Read original source ↗ Mistral AIReplay this moment in the explorer
Uncertainty marks flag where to check the sources; they are not proven causation.
Full source archive (current)
Current public reports; outside the historical replay.
First-party sources · 1 reports
- Mistral AI · Oct 20, 20:00Mistral AI launches Voxtral 3, a multilingual voice‑agent SDK with real‑time speech‑to‑text and text‑to‑speech models
Mistral AI has released Voxtral 3, a new family of speech‑to‑text and text‑to‑speech models that support more than 50 languages and dialects. The Voxtral Small Transcribe 3 model delivers highly accurate multilingual transcription with diarization and end‑of‑turn detection, while Voxtral Mini TTS 3 offers expressive, emotion‑adaptive speech synthesis. The accompanying SDK stitches listening, reasoning, tool‑calling, and speaking into a single real‑time streaming pipeline, enabling contact‑center agents, virtual assistants, and hands‑free copilots to perform multi‑step tasks across integrated systems. Voxtral 3 is available via Mistral’s EU‑hosted API or for sovereign self‑deployment, giving organizations control over data sovereignty and low‑latency interactions.
Read original source ↗
Independent coverage · 1 reports
- The Decoder · Oct 21, 01:17Mistral AI launches Voxtral 3 real-time voice agent stack
Mistral AI released Voxtral 3, a family of language models for real-time voice agents that transcribe speech in over 50 languages and generate speech in 35 languages with 50 dialects. The system includes a speech-to-text model, a text-to-speech model, and a voice agent SDK built on LiveKit, allowing agents to use external tools during live conversations. Users can deploy the system on their own infrastructure or via EU-hosted APIs.
Read original source ↗
Different reports do not necessarily disagree. Check original sources and publication dates.
Report timeline
Explore every side through dated source reports.
- The DecoderMistral AI launches Voxtral 3 real-time voice agent stack
Mistral AI released Voxtral 3, a family of language models for real-time voice agents that transcribe speech in over 50 languages and generate speech in 35 languages with 50 dialects. The system includes a speech-to-text model, a text-to-speech model, and a voice agent SDK built on LiveKit, allowing agents to use external tools during live conversations. Users can deploy the system on their own infrastructure or via EU-hosted APIs.
- Mistral AISelectedMistral AI launches Voxtral 3, a multilingual voice‑agent SDK with real‑time speech‑to‑text and text‑to‑speech models
Mistral AI has released Voxtral 3, a new family of speech‑to‑text and text‑to‑speech models that support more than 50 languages and dialects. The Voxtral Small Transcribe 3 model delivers highly accurate multilingual transcription with diarization and end‑of‑turn detection, while Voxtral Mini TTS 3 offers expressive, emotion‑adaptive speech synthesis. The accompanying SDK stitches listening, reasoning, tool‑calling, and speaking into a single real‑time streaming pipeline, enabling contact‑center agents, virtual assistants, and hands‑free copilots to perform multi‑step tasks across integrated systems. Voxtral 3 is available via Mistral’s EU‑hosted API or for sovereign self‑deployment, giving organizations control over data sovereignty and low‑latency interactions.