Feature: Add SGLang support for GRPO Trainer #3627
Draft
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
|
Testing result will be this comment:
More testing ongoing |
Collaborator
|
shall we close the other PR? |
zkx06111
reviewed
Jun 22, 2025
| requests; python_version < "3.13" | ||
| uvicorn; python_version < "3.13" | ||
|
|
||
| sglang = |
There was a problem hiding this comment.
Should this be sglang[all]>=0.4.6post2?
Without the [all], I'm getting lots of modulenotfounderrors.
zkx06111
reviewed
Jun 22, 2025
Comment on lines
+256
to
+262
| # start sglang-server | ||
| python3 -m sglang.launch_server --model-path qwen/qwen2.5-7b-instruct | ||
|
|
||
| # run "export CUDA_VISIBLE_DEVICES" | ||
| # run script | ||
| python3 grpo_test.py | ||
| ``` |
zkx06111
reviewed
Jun 22, 2025
docs/source/grpo_trainer.md
Outdated
| from trl import GRPOConfig, GRPOTrainer | ||
|
|
||
|
|
||
| dataset = load_dataset("trl-lib/tldr", split="train[:10%]”) |
There was a problem hiding this comment.
split="train[:10%]” should be split="train[:10%]"
Turn to the online server API Usage add test and fix bugs in result parsing Pass First test with fixing _update_sglang_weights Remove checkpoints from tracking and add to .gitignore config to run on single gpu successfully Update code to align with vllm save model and update weight save model only main process A runnable update_from_tensor version fix performance issue resolve comment: help strings resolve comment: help strings Update trl/trainer/grpo_config.py Update trl/trainer/grpo_config.py Update trl/trainer/grpo_config.py Update trl/trainer/grpo_config.py Update trl/trainer/grpo_config.py Update trl/trainer/grpo_config.py Update trl/trainer/grpo_trainer.py call raise_for_status remove duplicate doc string formatting add sglang to extras formatting import requests only when sglang is available formatting undo formatting undo formatting more undo last one! add initial docs add sglang last one now new line delete test scripts Update setup.cfg Update setup.cfg intiial sglang-serve cli script Update trl/trainer/grpo_trainer.py remove dead code Co-authored-by: Kashif Rasul <[email protected]> debug GRPO trainer change num_processes update how to run sglang
PrinsYin
force-pushed
the
sglang-server-rebase
branch
from
da2d521 to
d2ed28b
Compare
8 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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 re-implements and rebases the original work from #3370, which adds SGLang support as a rollout engine for GRPO training in TRL.
We have rebased the original PR onto the latest main branch and testing the behavior of weight updates.
See also #103
@ChangyiYang
Fixes # (issue)
Before submitting
Pull Request section?
to it if that's the case.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.