Voxtral Mini Transcribe Realtime

Name: Voxtral Mini Transcribe Realtime
Author: Mistral

Mistral · Released Feb 4, 2026 · Open weights

Low-latency streaming speech-to-text model for live transcription and voice agents

Open in the timeline

BENCHMARKS

Not in the Epoch Capabilities Index. Epoch benchmarks what it can run, so this is a gap in coverage, not a poor result.

Mistral's past year

15 release days in the past year, about one every 24 days. The next is expected around Nov 13, 2026. Forecasts ↗

NovDecJanFebMarAprMayJunJulAugSepOctNov

Specifications

RELEASE DATEFeb 4, 2026

WEIGHTSOpen weights

CONTEXT WINDOWNot listed

MAX OUTPUT0

REASONINGNot listed

TOOL CALLINGNot listed

INPUTSaudio/pcm, audio/pcmu, audio/pcma

OUTPUTStext

Models.dev record ↗

Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.