The Local LLM Index / Quantization & Formats / #221
bytedance/ABQ-LLM
by bytedance · Quantization & Formats · updated 1y ago
An acceleration library that supports arbitrary bit-width combinatorial quantization operations
27
momentum
247
stars
22
forks
#221
rank
cudallm-inferencemlsysquantized-networksresearch
View on GitHub →