Fix dequant mixed#4675
Open
irexyc wants to merge 7 commits into
Open
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR fixes dequantization behavior for compressed-tensors int4 weights in TurboMind, enabling successful loading/inference for the referenced Qwen3.5 AWQ-4bit checkpoint when mixed-format fusion groups require dequantization to a trivial (FP) format.
Changes:
- Adjust
_zeros_int4_symmetricto synthesize zero-points usingscales.dtype(avoids uint8 arithmetic wraparound during dequant). - Implement
CompressedTensorFormat.dequant()to supportdequant_mixed()flows for mixed-format fusion groups. - Update
_dequant_lineardocstring to reflect thatCompressedTensorFormat.dequantnow performs real work.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
lmdeploy/turbomind/weight_format.py |
Fixes synthesized zero-point dtype and adds CompressedTensorFormat.dequant() for mixed dequant paths. |
lmdeploy/turbomind/linear.py |
Updates dequantization documentation to include CompressedTensorFormat. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Motivation
fix https://huggingface.co/cyankiwi/Qwen3.5-27B-AWQ-4bit loading and inference