OpenMOSS-Team · First MLX build 31 Mar 2026
MOSS-Audio-Tokenizer
MLX
Text apache-2.0
A 22M parameter embeddings model from OpenMOSS-Team, with 2 MLX builds at 8-bit.
- Parameters
- 22M From Hugging Face
- Quantisation
- Unstated MOSS-Audio-Tokenizer-Nano
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 44 MB Weights, estimated
- Downloads
- 307 This build, last 30 days
- Task
- Embeddings Text
- MLX builds
- 2 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/MOSS-Audio-Tokenizer-NanoMLX builds
2 on mlx-community| MOSS-Audio-Tokenizer-Nano | Unstated | 44 MB | 3 GB | 307 |
|---|---|---|---|---|
| MOSS-Audio-Tokenizer-MLX-8bit | 8-bit | — | — | 0 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.