Loading the catalogue…
Mistral AI
voxtral-mini-tts-3
Released 20 Oct 2026
€12.82per 1M characters
Loading the catalogue…
| Inference Provider | Sovereignty | Runs in | Price | Status | Pin |
|---|---|---|---|---|---|
| L1 | €12.82 per 1M characters | Active |
8 use cases classified by the curator. The note says why.
The model name identifies it as Voxtral Mini TTS, a text-to-speech model; speech synthesis is its directly indicated use case.
A text-to-speech model can provide spoken output for voice agents, but no tool-calling, dialogue, or speech-recognition capabilities are documented.
Browser automation requires tool use and action planning, neither of which is documented for this model.
The TTS specialization can vocalize chatbot responses, but no conversational language, reasoning, or interaction capabilities are documented.
No function-calling support is flagged, while the model's stated specialization is text-to-speech.
No tool-calling or agent orchestration capability is flagged, and the model is identified as a TTS model.
The profile indicates text-to-speech rather than speech-to-text, and no transcription capability is flagged.
No voice-cloning capability or speaker-identity controls are documented; the available signal only supports generic speech synthesis.
Sourced numbers, not our own evaluations. Percentiles are against the speech models on Stav.
No benchmark source yet
curl https://api.stav.ai/v1/audio/speech \
-H "Authorization: Bearer $STAV_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "mistral-ai/voxtral-mini-tts-3", "input": "Hei, verden!", "voice": "alloy"}' \
--output speech.mp3→ X-Stav-Request-Id · X-Stav-Model · X-Stav-Provider → X-Stav-Sovereignty-Level — the rung the request was served at