In quantized sdpa dequant v #10097
Merged
In quantized sdpa dequant v #10097
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
Summary: for softmax(q @ k) @ v matmul, instead of leveraging our own kernel, dequant v and then leverage blas This improves prefill to 121 from 80 but degrades decode. I think I just need to implement implement gemm version of fp32_int8 matmul Reviewed By: metascroy Differential Revision: D71833063
kimishpatel
requested review from
jackzhxng,
iseeyuan,
larryliu0820 and
swolchok
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/10097
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit 27351af with merge base feead18 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D71833063 |
kimishpatel
added
the
release notes: examples
digantdesai
approved these changes
Apr 11, 2025
kirklandsign
pushed a commit
that referenced
this pull request
Apr 11, 2025
Differential Revision: D71833063 Pull Request resolved: #10097
keyprocedure
pushed a commit
to keyprocedure/executorch
that referenced
this pull request
Apr 21, 2025
Differential Revision: D71833063 Pull Request resolved: pytorch#10097
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Outdated suggestions cannot be applied.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.
Summary:
for softmax(q @ k) @ v matmul, instead of leveraging our own kernel, dequant v and then leverage blas
This improves prefill to 121 from 80 but degrades decode.
I think I just need to implement implement gemm version of fp32_int8 matmul
Reviewed By: metascroy
Differential Revision: D71833063