PyPI page
Home page
Author:
Tianyi Zhang
Summary:
DFloat11: Fast and memory-efficient GPU inference for losslessly compressed LLMs and diffusion models
Latest version:
0.5.0
Required dependencies:
accelerate
|
dahuffman
|
huggingface-hub
|
safetensors
|
tqdm
|
transformers
Optional dependencies:
cupy-cuda11x
|
cupy-cuda12x
Downloads last day:
50
Downloads last week:
506
Downloads last month:
2,288