Fix MXFP4 quantizer validation to allow CPU inference with dequantize option #39953
Conversation
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
Move dequantize check before CUDA availability check to allow CPU inference when quantization_config.dequantize is True. This enables users to run MXFP4 models on CPU by automatically converting them to BF16 format.
|
[For maintainers] Suggested jobs to run (before merge) run-slow: mxfp4 |
MekkCyber
approved these changes
Aug 6, 2025
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
SunMarc
added
the
for patch
ArthurZucker
pushed a commit
that referenced
this pull request
Aug 13, 2025
… option (#39953) * Fix MXFP4 quantizer validation to enable CPU dequantization Move dequantize check before CUDA availability check to allow CPU inference when quantization_config.dequantize is True. This enables users to run MXFP4 models on CPU by automatically converting them to BF16 format. * Add tests for MXFP4 quantizer CPU dequantization validation * fix: format mxfp4 test file with ruff
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.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.
What does this PR do?
This PR fixes a bug that prevented MXFP4 models from running on CPU when
quantization_config.dequantize=Truewas set.Problem
The validation logic in
Mxfp4HfQuantizerchecked CUDA availability before checking thedequantizeflag, causing failures on CPU-only environments even when dequantization was enabled.Solution
Reordered validation checks to prioritize
dequantizeconfiguration:dequantizeis enabled - if yes, skip GPU validationsChanges Made
dequantizecheck before CUDA validation inquantizer_mxfp4.pydequantize=TrueBefore submitting
Who can review?
@SunMarc @MekkCyber