feat(trainer): enable torch.compile by default#2489
Open
samsja wants to merge 1 commit into
Open
Conversation
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit f95f67a. Configure here.
Default `ModelConfig.compile` to `CompileConfig()` so transformer blocks are compiled by default. Configs that explicitly set `[model.compile]` continue to override as before. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
samsja
force-pushed
the
feat/enable-torch-compile-by-default
branch
from
f95f67a to
0f2fc1f
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
ModelConfig.compiletoCompileConfig()instead ofNone, soapply_compileruns by default and each transformer block is wrapped withtorch.compile.fullgraphstill defaults toFalse. Configs that explicitly set[model.compile](e.g.configs/nemotron_4node/rl.toml,configs/hendrycks_math/sanity.toml) continue to override as before.🤖 Generated with Claude Code
Note
Medium Risk
Changes default training behavior by enabling
torch.compileunless explicitly disabled, which can impact performance, debugging, and runtime compatibility for some models/environments.Overview
Enables model compilation by default by changing
ModelConfig.compile’s default fromNonetoCompileConfig(), so configurations that don’t specifymodel.compilewill now run the model throughtorch.compileautomatically.Existing configs that explicitly set
model.compilecontinue to override this default.Reviewed by Cursor Bugbot for commit 0f2fc1f. Bugbot is set up for automated code reviews on this repo. Configure here.