MLX Community · First MLX build 4 Jan 2026
Soprano-80M
MLX
Audio apache-2.0
A 94M parameter text to speech model, with 5 MLX builds, from 4-bit to bf16.
- Parameters
- 94M From Hugging Face
- Quantisation
- bf16 Soprano-80M-bf16
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 217 MB Weights, estimated
- Downloads
- 1.8K This build, last 30 days
- Task
- Text to speech Audio
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at bf16
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/Soprano-80M-bf16MLX builds
5 on mlx-community| Soprano-80M-4bit | 4-bit | 56 MB | 3 GB | 30 |
|---|---|---|---|---|
| Soprano-80M-5bit | 5-bit | 68 MB | 3 GB | 10 |
| Soprano-80M-6bit | 6-bit | 79 MB | 3 GB | 20 |
| Soprano-80M-8bit | 8-bit | 103 MB | 3 GB | 14 |
| Soprano-80M-bf16 | bf16 | 217 MB | 3 GB | 1.8K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.