PyPI page
Home page
Author:
Humming contributors
Summary:
JIT-compiled GPU GEMM kernels for quantized inference
Latest version:
0.1.16
Required dependencies:
cuda-bindings
|
jinja2
|
numpy
|
safetensors
|
tabulate
|
torch
|
tqdm
|
triton
Optional dependencies:
nvidia-cuda-cccl
|
nvidia-cuda-cccl-cu12
|
nvidia-cuda-nvcc
|
nvidia-cuda-nvcc-cu12
|
nvidia-cuda-nvrtc
|
nvidia-cuda-nvrtc-cu12
|
nvidia-cuda-runtime
|
nvidia-cuda-runtime-cu12
Downloads last day:
82,667
Downloads last week:
480,561
Downloads last month:
1,938,489