search

intel / auto-round

Advanced Quantization Algorithm for LLMs. This is official implementation of "Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs"
https://arxiv.org/abs/2309.05516
Apache License 2.0
132 stars 18 forks source link

support marlin in auto_round format #172

Closed wenhuach21 closed 2 weeks ago

wenhuach21 commented 1 week ago

regression in this pr