ResembleAI · First MLX build 17 Dec 2025
chatterbox-turbo
MLX
Audio apache-2.0
A 626M parameter text to speech model from ResembleAI, with 7 MLX builds, from 4-bit to fp16.
- Parameters
- 626M From Hugging Face
- Quantisation
- 4-bit chatterbox-turbo-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 415 MB Weights, estimated
- Downloads
- 240K This build, last 30 days
- Task
- Text to speech Audio
- MLX builds
- 7 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/chatterbox-turbo-4bitMLX builds
7 on mlx-community| chatterbox-turbo-4bit | 4-bit | 415 MB | 3 GB | 240K |
|---|---|---|---|---|
| chatterbox-turbo-5bit | 5-bit | 487 MB | 3 GB | 80 |
| chatterbox-turbo-6bit | 6-bit | 560 MB | 3 GB | 134 |
| chatterbox-turbo-8bit | 8-bit | 706 MB | 3 GB | 1.5K |
| Chatterbox-Turbo-TTS-4bit | 4-bit | 811 MB | 3 GB | 249 |
| Chatterbox-Turbo-TTS-8bit | 8-bit | 962 MB | 4 GB | 193 |
| Chatterbox-Turbo-TTS-fp16 | fp16 | 1.2 GB | 4 GB | 237 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.