Skip to content

Voxtral-4B-TTS-2603-mlx-bf16

mlx-community/Voxtral-4B-TTS-2603-mlx-bf16
bf16 Open on Hugging Face
Parameters
—
Weights (est.)
—
Mac memory (est.)
—

Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.

Details

Base model
mistralai/Voxtral-4B-TTS-2603 (finetune)
Vendor
Mistral
Task
Text to speech
Quantisation
bf16
Licence
cc-by-nc-4.0
Downloads
20K in the last 30 days · 27K all time · 1 likes
Published
2026-03-26 · updated 2026-03-27 (7mo ago)
Languages
en, fr, es, de, it, pt, nl, ar, hi
Tags
voxtral_tts, speech, speech generation

Other MLX builds of mistralai/Voxtral-4B-TTS-2603

Build Quantisation Weights Mac memory Downloads
Voxtral-4B-TTS-2603-mlx-4bit 4-bit ≈ 2.5 GB 8 GB+ 3.1K
Voxtral-4B-TTS-2603-mlx-6bit 6-bit ≈ 3.5 GB 8 GB+ 1.8K
Voxtral-4B-TTS-2603-mlx-bf16 (this one) bf16 — — 20K