VPTQ, A Flexible and Extreme low-bit quantization algorithm

AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:

A profiler to disclose and quantify hardware features on GPUs.

A list of awesome compiler projects and papers for tensor computation and deep learning.

分享计算机视觉每天的arXiv文章

String splitting benchmarks

Classical equations and diagrams in machine learning