The Local LLM Index / Quantization & Formats / #226
bytedance/ABQ-LLM
by bytedance · Quantization & Formats · updated 2y ago
An acceleration library that supports arbitrary bit-width combinatorial quantization operations
27
momentum
249
stars
24
forks
#226
rank
cudallm-inferencemlsysquantized-networksresearch
View on GitHub →