---
id: "mistralai/voxtral-small-transcribe-3-0"
also: ["mistralai/voxtral-small-transcribe-3-0-20261020"]
card_usd_per_m: "73.3333 in, 0.0 out, at mistral/eu"
cheapest_list_usd_per_m: "66.6667 in, 0.0 out, at mistral/zdr on openrouter"
cheapest_effective_usd_per_m: "143.5711 in, 0.0 out, at mistral/zdr on openrouter"
context: 65536
max_output: 52428
input: ["audio"]
output: ["transcription"]
released: "2026-10-20"
api: ["/api/v1/audio/transcriptions"]
endpoints: 3
gateways: "openrouter (3)"
no_training: 2
trains: 0
retains_prompts: 0
---
# Mistral: Voxtral Small Transcribe 3

Voxtral Small Transcribe 3 is Mistral's flagship speech-to-text model. It takes audio input and returns a transcript with timestamps and speaker diarization, and supports context biasing for domain-specific vocabulary. Suited...