Mistral releases Voxtral 3 speech models and a voice agent SDK: a 4B transcriber at $0.0035 a minute, TTS at $13 per million characters
A real-time pipeline that listens, reasons, calls tools and speaks, available through Mistral's EU-hosted API or on your own infrastructure.
// Key points
- The transcription model Voxtral Small Transcribe 3 covers 50+ languages with diarization, context biasing and end-of-turn detection; Mistral says it outperforms Whisper, Scribe v2 and Gemini 3.5 Transcribe.
- The Mini transcriber has 4B parameters and costs $0.0035 per minute; Voxtral Realtime Transcribe 3 runs at under 500 ms latency in 35+ languages.
- The TTS model Voxtral Mini TTS 3.0 supports 35+ languages and 50+ dialects at $13 per million characters; Pro Voice reproduces a speaker's style from about 30 minutes of audio.
- A new voice agent SDK combines listening, reasoning, tool calling and speaking in one real-time streaming pipeline.
Builder's takeFor voice-first products like AI Interview, what I care about most is end-of-turn detection and self-deployment: a bad interruption call wrecks the interview, and candidate recordings should ideally stay in my own environment. I'll first compare Chinese error rate and end-to-end latency against my current transcription on a batch of real interview recordings before deciding to switch.
// Background · from #AI agents
Full timeline →- Oct 21 Amazon Bedrock adds structured outputs for all Claude models: JSON Schema output and strict tool use
- Oct 21 IBM's ALTK-Evolve appends learned rules to the prompt, taking a GPT-4.1 support agent to 67.5% Pass^4 on τ²-bench retail without touching weights
- Oct 20 Cloudflare releases Clairic, an extraction model that returns typed values in one forward pass, billed on input tokens only, in preview on Workers AI