Skip to content

Voxtral Mini Transcribe Realtime

mistral/voxtral-mini-transcribe-realtime

Voxtral Mini Transcribe Realtime (February 2026), a 4B streaming speech-to-text model with Apache 2.0 open weights, 13 languages and a configurable delay from 240 ms to 2.4 s; at 480 ms Mistral puts it within 1-2% WER of its batch model.

API
import WebSocket from "ws";
import { readFile } from "node:fs/promises";

// Use mono, 16 kHz, signed 16-bit little-endian PCM.
const audio = await readFile("recording.pcm");
const url = new URL("/v1/audio/transcriptions/stream", process.env.AUDIOLAB_API_URL);
url.protocol = url.protocol === "https:" ? "wss:" : "ws:";
const socket = new WebSocket(url, {
  headers: { Authorization: `Bearer ${process.env.AUDIOLAB_API_KEY}` },
});
socket.on("open", () => socket.send(JSON.stringify({
  type: "session.start", model: "mistral/voxtral-mini-transcribe-realtime",
  encoding: "pcm_s16le", sample_rate: 16000, channels: 1,
})));
socket.on("message", (data) => {
  const event = JSON.parse(data.toString());
  console.log(event);
  if (event.type === "session.started") {
    for (let offset = 0; offset < audio.length; offset += 3200)
      socket.send(audio.subarray(offset, offset + 3200));
    socket.send(JSON.stringify({ type: "session.finish" }));
  }
});
Input Price
$0.0051 / minute
Output Price
—
Input
Audio
Output
Text

Capabilities

Explore what Voxtral Mini Transcribe Realtime supports, then use the API section for a request tailored to this model.

Stream

Languages

Multilingual

Model Information

Provider
Mistral AI
Product
Speech-to-Text
Release Date
Feb 4, 2026
Data Policy
Configurable
Region
Global
Input
Audio
Output
Text
Provider Documentation

Playground

Try Voxtral Mini Transcribe Realtime by Mistral AI. Requests use the selected workspace’s credits and API pricing.

Voxtral Mini Transcribe RealtimeCredits —
Voxtral Mini Transcribe Realtime
This model uses a dedicated input format. Explore its API example or open the supported playground.

Providers

Compare available endpoints, pricing, and context limits for Voxtral Mini Transcribe Realtime.

ProviderContextOutputInput PriceOutput PriceEndpoint
Mistral AINot specifiedText$0.0051 / minute—
mistral/voxtral-mini-transcribe-realtime:mistral

Prices are catalog rates. Your workspace’s estimate and endpoint pricing determine the final charge.

Input and Output Limits

Formats and limits AudioLab enforces on each request.

LimitValue
Max bytesNot specified
Max secondsNot specified
Formats

Provider Notes

Voxtral Mini Transcribe 2 is the current pinned dedicated uploaded-audio model; voxtral-mini-transcribe is the mutable latest alias. Voxtral Mini Transcribe Realtime is a separate WebSocket-only model with target streaming delay control and separate pricing. Voxtral Small is a prompted audio+text chat model, not the dedicated transcription endpoint: it accepts MP3/WAV and does not expose timestamps, diarization, context bias, or realtime transcription. The deprecated Voxtral Mini 2507 API route is intentionally excluded.

Performance

Status

Current catalog availability. Your workspace’s access is checked separately before a request runs.

Getting Started

Use your workspace API key and the exact model identifier below. Set the API URL to the endpoint shown in your workspace and keep credentials on your server.

API
import WebSocket from "ws";
import { readFile } from "node:fs/promises";

// Use mono, 16 kHz, signed 16-bit little-endian PCM.
const audio = await readFile("recording.pcm");
const url = new URL("/v1/audio/transcriptions/stream", process.env.AUDIOLAB_API_URL);
url.protocol = url.protocol === "https:" ? "wss:" : "ws:";
const socket = new WebSocket(url, {
  headers: { Authorization: `Bearer ${process.env.AUDIOLAB_API_KEY}` },
});
socket.on("open", () => socket.send(JSON.stringify({
  type: "session.start", model: "mistral/voxtral-mini-transcribe-realtime",
  encoding: "pcm_s16le", sample_rate: 16000, channels: 1,
})));
socket.on("message", (data) => {
  const event = JSON.parse(data.toString());
  console.log(event);
  if (event.type === "session.started") {
    for (let offset = 0; offset < audio.length; offset += 3200)
      socket.send(audio.subarray(offset, offset + 3200));
    socket.send(JSON.stringify({ type: "session.finish" }));
  }
});
Read the Endpoint Reference

Request Parameters

Available parameters depend on the provider and request format. The endpoint reference explains accepted values and required fields.

ParameterDetails
modelmistral/voxtral-mini-transcribe-realtime — Required model identifier.
stream (completed-audio SSE)View Accepted Values
target_streaming_delay_ms (Voxtral Mini Transcribe 2 realtime)View Accepted Values

Input and Output Limits

Formats and limits AudioLab enforces on each request.

LimitValue
Max bytesNot specified
Max secondsNot specified
Formats

Similar Models

Explore more speech-to-text and compare their capabilities.

ModelContextInput PriceOutput Price
Voxtral Mini Transcribe 2Mistral AINot specified$0.0026 / minute—
Voxtral Mini Transcribe 2Mistral AINot specified$0.0026 / minute—
Voxtral SmallMistral AINot specified$0.0034 / minute—
MAI-Transcribe 2Microsoft AzureNot specified$0.0014 / minute—
Gemini 3.5 TranscribeGoogle GeminiNot specified$0.0043 / minute—
Saaras v4Sarvam AINot specified₹0.425 / minute—
View All Speech-to-Text