Mistral’s 4B TTS. It uses its built-in voices and not a cloned reference clip.
TTS
Voxtral TTS
Voxtral TTS is Mistral’s 4B local text-to-speech model with 20 voices across 9 languages for OpenToys.
Documentation Index
Fetch the complete documentation index at: /docs/llms.txt
Use this file to discover all available pages before exploring further.
Voxtral TTS is Mistral’s 4B local text-to-speech model with 20 voices across 9 languages for OpenToys.
| Model | mlx-community/Voxtral-4B-TTS-2603-mlx-4bit |
| Memory | About 5 GB RAM, 3 GB on disk |
| Quality | Natural |
| Voices | 20 across 9 languages |
| License | Non-commercial |