MLX Community · First MLX build 19 Feb 2026
kitten-tts-nano-0.8
MLX
Audio apache-2.0
A 15M parameter text to speech model, with 7 MLX builds, from 4-bit to bf16.
- Parameters
- 15M From Hugging Face
- Quantisation
- Unstated kitten-tts-nano-0.8
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 58 MB Weights, estimated
- Downloads
- 112 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-nano-0.8MLX builds
7 on mlx-community| kitten-tts-nano-0.8-4bit | 4-bit | 30 MB | 3 GB | 17 |
|---|---|---|---|---|
| kitten-tts-nano-0.8-5bit | 5-bit | 31 MB | 3 GB | 20 |
| kitten-tts-nano-0.8-6bit | 6-bit | 32 MB | 3 GB | 8 |
| kitten-tts-nano-0.8-8bit | 8-bit | 38 MB | 3 GB | 83 |
| kitten-tts-nano-0.8 | Unstated | 58 MB | 3 GB | 112 |
| kitten-tts-nano-0.8-fp16 | fp16 | 58 MB | 3 GB | 25 |
| kitten-tts-nano-0.8-bf16 | bf16 | 58 MB | 3 GB | 6 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.