PyPI page
Home page
Author:
Manjunath Janardhan
Summary:
Extreme weight and KV cache compression for LLMs on Apple Silicon (MLX implementation of Google's TurboQuant)
Latest version:
0.28.1
Required dependencies:
mlx
|
mlx-lm
|
numpy
|
transformers
Optional dependencies:
blobfile
|
build
|
datasets
|
mlx-vlm
|
pytest
|
tiktoken
|
transformers
|
twine
Downloads last day:
86
Downloads last week:
153
Downloads last month:
574
7-day smoothing