Switch FP8 per tensor quant to use torch._scaled_mm#44864
Closed
Switch FP8 per tensor quant to use torch._scaled_mm#44864
torch._scaled_mm#44864Conversation
|
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. |
Member
Author
|
Closing in favor of #44895 |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 switches FP8 per-tensor implementation to rely on the official torch impl
torch._scaled_mm. Note thattorch._scaled_mmdon't explicitly support per tensor. We hack the api a bit as it only support per row.It is actually faster than the hotfix we did with the existing triton kernels that were used for block quantization (128,128).
Will add some benchmark.