Mistral Vox
by Mistral AI
· Vox family · released Sep 25, 2025
Closed
Active
Mistral's speech understanding model that transcribes audio with natural punctuation, speaker diarization, and native summarization/translation. It beats OpenAI Whisper on French and English audio benchmarks.
No pricing recorded for this model yet.
Specifications
| Model type | Multimodal |
|---|---|
| Architecture | Unknown |
| Parameters | Not disclosed |
| Context window | Not disclosed tokens |
| Max output | Not disclosed tokens |
| Knowledge cutoff | Not disclosed |
| License | Proprietary |
| Modalities | Audio input Text output |
| Tags | - |
| End of life | Not disclosed |
| Official page | Link |
| Last verified | Not disclosed |