As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/voxtral-tts/

# Voxtral TTS

Mistral AI, Voxtral family. Status: current. Type: speech and audio. Released: 2026-03-23.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Mistral API | `voxtral-tts-2603` | https://api.mistral.ai/v1/audio/speech | https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03 |
| Hugging Face | `mistralai/Voxtral-4B-TTS-2603` |  | https://huggingface.co/mistralai/Voxtral-4B-TTS-2603 |
| Web app |  |  | https://chat.mistral.ai |

## Pricing

- Per 1K characters: $0.016
- Unit as published: USD per 1K characters ($16 per 1M) on Mistral API
- Source: https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03
- Last checked: 2026-09-29

## Specs

- Input: text, audio
- Output: audio
- Context window: not published
- Max output: not published
- Open weights: yes
- Licence: cc-by-nc-4.0

## Where it stands

- Newer speech and audio models in the registry since its release: 63. The 5 newest:
  - NVIDIA Nemotron 3.5 Transcribe (30B-A4B) (2026-10-28) https://postcutoff.com/m/nemotron-3-5-transcribe/
  - Hy-SimulMT-30B-A3B (2026-10-22) https://postcutoff.com/m/hy-simulmt-30b-a3b/
  - Eleven Realtime v1 Max / Mini (2026-10-20) https://postcutoff.com/m/eleven-realtime-v1-max/
  - Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-transcribe-3/
  - Voxtral Small Transcribe 3 (2026-10-20) https://postcutoff.com/m/voxtral-small-transcribe-3/
- Before it from Mistral AI: Voxtral Transcribe 2 (Mini Transcribe V2 + Voxtral Realtime) (2026-02-04) https://postcutoff.com/m/voxtral-transcribe-2/
- After it from Mistral AI: Voxtral TTS 3 (Voxtral Mini TTS 3) (2026-10-20) https://postcutoff.com/m/voxtral-mini-tts-3/
- Weights are public. Licence: CC BY-NC 4.0. See the licence for commercial terms.

## What stands out

- Zero-shot voice cloning from ~3 s: Clones a voice (accent, fillers, rhythm) from a few seconds of reference audio without a transcript; 68.4% human-preference win rate vs ElevenLabs Flash v2.5 on multilingual cloning (Mistral-reported). Source: https://mistral.ai/news/voxtral-tts
- Open-weight 4B TTS with low latency: 3.4B decoder + 390M flow-matching acoustic transformer + 300M codec; ~70 ms model latency (~90 ms time-to-first-audio via API), RTF ~9.7x, up to 2 min native generation. Source: https://mistral.ai/news/voxtral-tts

## Notes

Mistral's first TTS model. 9 languages: English, French, German, Spanish, Dutch, Portuguese, Italian, Hindi, Arabic. Weights are CC BY-NC 4.0 (non-commercial); commercial use via API. Docs model-card page shows id voxtral-tts-2603 on the overview (a 'voxtral-mini-tts-2603' alias also appears on the card). Succeeded by Voxtral TTS 3 (voxtral-mini-tts-3-0, 2026-10-20, API-only): see voxtral-mini-tts-3.

Call `POST https://api.mistral.ai/v1/audio/speech` with `model: voxtral-tts-2603` (see https://docs.mistral.ai/capabilities/audio/text_to_speech for the request schema and voice options).

Sources: https://mistral.ai/news/voxtral-tts · https://docs.mistral.ai/models/overview · https://docs.mistral.ai/models/model-cards/voxtral-tts-26-03

Model page: https://postcutoff.com/m/voxtral-tts/