Raise ValueError when nvfp4 pack tensor has odd number of columns #402
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
This raises an explicit error for a case that previously threw an implicit error during the reshape op at the end of the function. Signed-off-by: Fynn Schmitt-Ulms <[email protected]>
src/compressed_tensors/compressors/quantized_compressors/nvfp4_quantized.py
Show resolved
Hide resolved
brian-dellabetta
approved these changes
Jul 23, 2025
|
should probably validate on vllm |
|
I've compressed a llama3.1 8B model, using this change and successfully loaded it into vllm and generated a sentence. |
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Outdated suggestions cannot be applied.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.
Fixes #401 (option 3 as discussed w/ @brian-dellabetta)
Also introduces a new test for odd columned tensors.
Note: we check that the function raises either a
ValueError(current behavior) or atorch._dynamo.exc.Unsupported(expected behavior after #400 merges).