Skip to content

Add quantization support for bridge mode weight update#1564

Open
jairuigou wants to merge 1 commit intoTHUDM:mainfrom
jairuigou:quantization-support
Open

Add quantization support for bridge mode weight update#1564
jairuigou wants to merge 1 commit intoTHUDM:mainfrom
jairuigou:quantization-support

Conversation

@jairuigou
Copy link

I noticed that bridge mode does not yet support quantization during weight updates. So I made some changes:

  • Add quantization_config parameter to postprocess_hf_param() to support fp8 and compressed-tensors quantization
  • Update HfWeightIteratorBridge to pass quantization config during weight conversion

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant