Instructions to use Audante/MOSS-Music-8B-Instruct-mlx-5bit-mix8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Audante/MOSS-Music-8B-Instruct-mlx-5bit-mix8 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download Audante/MOSS-Music-8B-Instruct-mlx-5bit-mix8 --local-dir MOSS-Music-8B-Instruct-mlx-5bit-mix8
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
MOSS-Music-8B-Instruct (MLX, mixed 5/8-bit)
Derived from a copy of OpenMOSS-Team/MOSS-Music-8B-Instruct at revision fce7f8304e96cc2d3398b8106456cbb2ecec3139, by OpenMOSS. This is not an official release.
Changes
Converted to MLX and quantised (affine, group size 64) to 5 bit with some layers in 8 bit; the per-layer settings are in config.json. No retraining.
Licence
Apache License 2.0, as the source โ see LICENSE.
All rights to the model remain with its authors.
SHA256SUMS, where present, lists the checksum of every file.
- Downloads last month
- 97
Model size
9B params
Tensor type
U32
ยท
BF16 ยท
Hardware compatibility
Log In to add your hardware
5-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support
Model tree for Audante/MOSS-Music-8B-Instruct-mlx-5bit-mix8
Base model
OpenMOSS-Team/MOSS-Music-8B-Instruct