4-bit Open on Hugging Face
Parameters
623M
Weights (est.)
≈ 811 MB
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
- ResembleAI/chatterbox-turbo (finetune)
- Vendor
- ResembleAI
- Task
- Text to speech
- Quantisation
- 4-bit
- Licence
- —
- Downloads
- 238 in the last 30 days · 3.0K all time · 3 likes
- Published
- 2025-12-22 · updated 2025-12-22 (10mo ago)
- Languages
- en
- Tags
- mlx-audio-plus, chatterbox_turbo, chatterbox
Other MLX builds of ResembleAI/chatterbox-turbo
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| chatterbox-turbo-4bit | 4-bit | ≈ 415 MB | 8 GB+ | 240K |
| chatterbox-turbo-5bit | 5-bit | ≈ 487 MB | 8 GB+ | 79 |
| chatterbox-turbo-6bit | 6-bit | ≈ 560 MB | 8 GB+ | 131 |
| chatterbox-turbo-8bit | 8-bit | ≈ 706 MB | 8 GB+ | 1.5K |
| Chatterbox-Turbo-TTS-4bit (this one) | 4-bit | ≈ 811 MB | 8 GB+ | 238 |
| Chatterbox-Turbo-TTS-8bit | 8-bit | ≈ 962 MB | 8 GB+ | 183 |
| Chatterbox-Turbo-TTS-fp16 | fp16 | ≈ 1.2 GB | 8 GB+ | 225 |