MLX Community · First MLX build 19 Feb 2026
kitten-tts-micro-0.8
MLX
Audio apache-2.0
A 36M parameter text to speech model, with 7 MLX builds, from 4-bit to fp16.
- Parameters
- 36M From Hugging Face
- Quantisation
- Unstated kitten-tts-micro-0.8
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 133 MB Weights, estimated
- Downloads
- 28 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 Unstated
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/kitten-tts-micro-0.8MLX builds
7 on mlx-community| kitten-tts-micro-0.8-4bit | 4-bit | 106 MB | 3 GB | 12 |
|---|---|---|---|---|
| kitten-tts-micro-0.8-5bit | 5-bit | 107 MB | 3 GB | 9 |
| kitten-tts-micro-0.8-6bit | 6-bit | 109 MB | 3 GB | 6 |
| kitten-tts-micro-0.8-8bit | 8-bit | 111 MB | 3 GB | 10 |
| kitten-tts-micro-0.8 | Unstated | 133 MB | 3 GB | 28 |
| kitten-tts-micro-0.8-bf16 | bf16 | 133 MB | 3 GB | 18 |
| kitten-tts-micro-0.8-fp16 | fp16 | 133 MB | 3 GB | 11 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.