MLX Community · First MLX build 8 May 2025
Spark-TTS-0.5B
MLX
Audio cc-by-nc-sa-4.0
A 507M parameter text to speech model, with 3 MLX builds, from 6-bit to bf16.
- Parameters
- 507M From Hugging Face
- Quantisation
- bf16 Spark-TTS-0.5B-bf16
- Memory needed
- 4 GB Estimate, unified memory
- Download
- 1.0 GB Weights, estimated
- Downloads
- 1.5K This build, last 30 days
- Task
- Text to speech Audio
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 4 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/Spark-TTS-0.5B-bf16MLX builds
3 on mlx-community| Spark-TTS-0.5B-6bit | 6-bit | 412 MB | 3 GB | 44 |
|---|---|---|---|---|
| Spark-TTS-0.5B-8bit | 8-bit | 538 MB | 3 GB | 105 |
| Spark-TTS-0.5B-bf16 | bf16 | 1.0 GB | 4 GB | 1.5K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.