tech
February 4, 2026
Voxtral transcribes at the speed of sound.
Usage-based, starting around €5K/month for your size.
TL;DR
- Voxtral Transcribe 2 includes Mini Transcribe V2 for batch and Realtime for live applications.
- Models offer state-of-the-art transcription quality, speaker diarization, and ultra-low latency.
- Voxtral Mini Transcribe V2 provides context biasing and word-level timestamps in 13 languages.
- Voxtral Realtime features configurable latency down to sub-200ms for voice agents and real-time applications.
- Voxtral Realtime is open-weights under Apache 2.0 for edge deployment and privacy.
- Mistral Studio now includes an audio playground to test Voxtral Transcribe 2 features.
- Voxtral Mini Transcribe V2 boasts industry-leading accuracy at a low cost, outperforming competitors like GPT-4o mini and Gemini 2.5 Flash.
- Enterprise features include speaker diarization, context biasing, word-level timestamps, expanded language support, noise robustness, and longer audio support (up to 3 hours).
- Voxtral Realtime is priced at $0.006/min via API and available as open weights.
- Voxtral Mini Transcribe V2 is priced at $0.003/min via API.