# Mistral Voxtral 3: Transcription From $0.0035 a Minute and a Full Voice Agent API at $0.04 a Minute

Mistral now sells every layer of a voice agent: speech-to-text, real-time transcription, text-to-speech and a LiveKit-based SDK that strings them together with a Mistral model. Here are the prices and what the benchmarks do and do not show.

By Priya Nair (AI Industry Reporter) — published 2026-10-22, updated 2026-10-22
Source: https://aiscoutdaily.com/news/mistral-voxtral-3-voice-agent-sdk-pricing

## TL;DR
Mistral released Voxtral 3 on 20 October 2026, a family of speech models plus a voice agent API. Voxtral Small Transcribe 3 costs $0.004 per minute, Voxtral Mini Transcribe 3 costs $0.0035 per minute in batch, the real-time Mini model costs $0.008 per minute with sub-500ms latency, and Voxtral Mini TTS 3 costs $13 per million characters. The complete voice agent API costs $0.04 per minute with Mistral Small 4 or $0.08 per minute with a larger model, and its SDK is built on LiveKit with SIP and WebRTC support. Mistral's own FLEURS chart puts Small Transcribe 3 at 4.1% word error rate, level with Microsoft and ahead of Gemini, ElevenLabs Scribe and OpenAI, but those are vendor-run numbers.

Until now, a team building a phone or voice agent on Mistral had to bring its own speech layers. Voxtral 3 changes that. Mistral now offers transcription, real-time transcription, speech synthesis and an agent SDK under one bill, and it priced the full loop per minute rather than per token.

## The price list

| Model or service | Price | What it is for |
| --- | --- | --- |
| Voxtral Small Transcribe 3 | $0.004 per minute | Highest-accuracy transcription: 50+ languages, speaker diarization, context biasing, end-of-turn detection |
| Voxtral Mini Transcribe 3 | $0.0035 per minute (batch) | Cheaper 4B-parameter transcription for files |
| Voxtral Mini Transcribe Realtime 3 | $0.008 per minute | Live transcription under 500ms |
| Voxtral Mini TTS 3 | $13 per 1M characters | Speech in 35+ languages and 50+ dialects, presets in 18 languages, Pro Voice cloning from 30 minutes of audio |
| Voice agent API, Mistral Small 4 | $0.04 per minute | Listen, think and speak in one call |
| Voice agent API, larger Mistral model | $0.08 per minute | Same loop with a stronger reasoning model |

*Mistral's Voxtral 3 announcement, 20 October 2026. All available through EU, US and global API endpoints, or self-deployed.*

One small inconsistency: Mistral's page gives the real-time model's language count as 35+ in one place and 40+ in another. Test your target languages before you commit.

## The agent SDK

- Built on LiveKit, the open-source real-time audio framework many voice agent teams already use.
- Supports SIP for phone lines and WebRTC for browser and app calls.
- Mistral says a full turn, from the caller finishing a sentence to the agent starting its reply, completes in under one second.
- End-of-turn detection in the transcription models is what lets the agent know when to speak without cutting people off.

## The benchmarks, as Mistral reports them

| Model | FLEURS-45 word error rate (lower is better) |
| --- | --- |
| Voxtral Small Transcribe 3 | 4.1% |
| Microsoft AI Transcribe 2 | 4.1% |
| Voxtral Mini Transcribe 3 | 4.6% |
| ElevenLabs Scribe v2 | 4.9% |
| Gemini 3.5 Transcribe | 6.1% |
| GPT Transcribe | 6.3% |
| Whisper large-v3 | 16.0% |

*Mistral's own chart, forced-language setting. Not independently verified.*

Mistral also reports an 18.4% concatenated minimum-permutation word error rate for diarization on Small Transcribe 3, and keyphrase recall of 94.6% when you pass it context terms against 81.4% without. That second number is the practical one: if your calls are full of product names or medical terms, feeding them in as context matters more than the model choice.

## What the coverage left out

- Weights: earlier Voxtral releases shipped open weights. The Voxtral 3 announcement mentions self-deployment but does not state an open-weights licence for each model, so confirm before planning an on-premises build.
- Benchmarks: every number above comes from Mistral. Microsoft ties Small Transcribe 3 on the headline chart, so the accuracy lead is not as clear as some headlines suggest.
- Full price table: most write-ups quoted one or two prices. The agent API at $0.04 per minute is the figure to compare against a stitched-together stack of separate speech, model and voice vendors.

## What this means for teams building voice agents

- If you run high call volumes, $0.04 per minute for the whole loop is a simple number to model against your current per-minute cost across three vendors.
- If you already use LiveKit, trying Voxtral is mostly a provider swap rather than a rebuild.
- If data location matters, the EU endpoint is the main reason to look at Mistral over US-only providers.
- Run your own word error test on real recordings. Vendor charts on FLEURS rarely match noisy phone audio.

## Sources
- [Mistral AI: Voxtral 3 (official announcement, Oct 20, 2026)](https://mistral.ai/news/voxtral-3/)
- [The Decoder: Mistral AI launches Voxtral 3, a full voice agent stack that listens, thinks and talks in real time](https://the-decoder.com/mistral-ai-launches-voxtral-3-a-full-voice-agent-stack-that-listens-thinks-and-talks-in-real-time/)
- [TestingCatalog: Mistral launches Voxtral 3 with a voice agent SDK](https://www.testingcatalog.com/mistral-launches-voxtral-3-with-a-voice-agent-sdk/)