Voxtral-Mini-4B-Realtime-2602-4bit
mlx-community/Voxtral-Mini-4B-Realtime-2602-4bit
Speech recognition · mlx
4-bit Open on Hugging Face
Parameters
4.4B
Weights (est.)
≈ 3.1 GB
Mac memory (est.)
8 GB+
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-Mini-4B-Realtime-2602 (quantized)
- Vendor
- Mistral
- Task
- Speech recognition
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 2.3K in the last 30 days · 13K all time · 17 likes
- Published
- 2026-02-06 · updated 2026-02-10 (9mo ago)
- Languages
- ar, de, en, es, fr, hi, it, nl, pt, zh, ja, ko, ru
- Tags
- voxtral_realtime, mlx-audio, speech-to-text, streaming, realtime
Other MLX builds of mistralai/Voxtral-Mini-4B-Realtime-2602
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Voxtral-Mini-4B-Realtime-2602-4bit (this one) | 4-bit | ≈ 3.1 GB | 8 GB+ | 2.3K |
| Voxtral-Mini-4B-Realtime-6bit | 6-bit | ≈ 3.6 GB | 8 GB+ | 401 |
| Voxtral-Mini-4B-Realtime-2602-fp16 | fp16 | ≈ 8.9 GB | 16 GB+ | 456 |