Metadata-Version: 2.1
Name: quant-matmul
Version: 0.0.0
Summary: Quantized MatMul in CUDA with a PyTorch interface
Home-page: https://github.com/oadirt/quant-matmul
Author: Tri Dao
Author-email: tri@tridao.me
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: Unix
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENSE

# Quantized matmul in CUDA, with a PyTorch interface
Original code from FasterTransformer / TensorRT-LLM:
https://github.com/NVIDIA/TensorRT-LLM/tree/main/cpp/tensorrt_llm/kernels

Adapted to support a different quantization scheme.
