ModAI changes to export xnnpack delegated non_lowered_server_model #10989
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/10989
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 0ad1f99 with merge base b73f9d5 ( 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: D70704201 |
tarun292
added
topic: not user facing
release notes: none
facebook-github-bot
pushed a commit
that referenced
this pull request
May 19, 2025
…10989) Summary: This adds a xnnpack delegated model to non_lowered_server_model. This will help in speeding up server evals in aten mode by delegating to XNNPack. We run a const_prop_pass before running the delegation because this will help get rid of some unnecessary q=>dq patterns that will slow the model down. Improvements seen for some models inference time are: MLD model ~900ms=>450ms OFI model ~450ms => 230ms Reviewed By: navsud Differential Revision: D70704201
facebook-github-bot
force-pushed
the
export-D70704201
branch
from
95af8c4 to
aa7a1d4
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70704201 |
YIWENX14
approved these changes
May 20, 2025
Gasoonjia
approved these changes
May 20, 2025
…10989) Summary: This adds a xnnpack delegated model to non_lowered_server_model. This will help in speeding up server evals in aten mode by delegating to XNNPack. We run a const_prop_pass before running the delegation because this will help get rid of some unnecessary q=>dq patterns that will slow the model down. Improvements seen for some models inference time are: MLD model ~900ms=>450ms OFI model ~450ms => 230ms Reviewed By: navsud, YIWENX14, Gasoonjia Differential Revision: D70704201
facebook-github-bot
force-pushed
the
export-D70704201
branch
from
aa7a1d4 to
0ad1f99
Compare
|
This pull request was exported from Phabricator. Differential Revision: D70704201 |
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:
This adds a xnnpack delegated model to non_lowered_server_model. This will help in speeding up server evals in aten mode by delegating to XNNPack. We run a const_prop_pass before running the delegation because this will help get rid of some unnecessary q=>dq patterns that will slow the model down.
Improvements seen for some models inference time are:
MLD model ~900ms=>450ms
OFI model ~450ms => 230ms
Reviewed By: navsud
Differential Revision: D70704201