8-bit 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
- bosonai/higgs-tts-2-3b-base (finetune)
- Vendor
- bosonai
- Task
- Text to speech
- Quantisation
- 8-bit
- Licence
- apache-2.0
- Downloads
- 596 in the last 30 days · 1.8K all time · 2 likes
- Published
- 2026-04-18 · updated 2026-04-18 (6mo ago)
- Languages
- en, zh, ko, de, es
- Tags
- mlx-audio, higgs_audio, voice-cloning, higgs-audio
Other MLX builds of bosonai/higgs-tts-2-3b-base
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| higgs-audio-v2-3B-mlx-q8 (this one) | 8-bit | — | — | 596 |
| higgs-audio-v2-3B-mlx-q6 | 6-bit | — | — | 139 |