[CB] [Minor] Add parameter to tune default compile level#46533
Merged
Conversation
remi-or
changed the title
|
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. |
Contributor
|
CI Dashboard: View test results in Grafana |
ArthurZucker
approved these changes
Jun 12, 2026
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Currently, continuous batching exposes the parameter
use_default_compile_configsto the users so that they may enabletorch.compileeasily. But that parameter enables compile withmax-autotune-no-cudagraphsanddynamic=Falsefor best performances. This makes for a long warmup, which is not great for tests or to iterate over benchmarks.This PR cleanly deprecates this parameter and instead exposes
default_compile_levelwhich is an integer parameter that controls default compile behavior. Level 0 means no compile, and level 1-3 activate compile, with increasing performance and warmup duration.If a user passes the old parameter, a warning is raised, and the level is set to 3, so that the behaviour is the same as with the old parameter.
Performance
No behavioral change, so same perf.
Tests
TBD
Review
Reviewed by claude + coumpound engineering.