feat: use async vllm engine (only used in unit tests) #418
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
parthchadha
added
CI:L0
Signed-off-by: Parth Chadha <[email protected]>
parthchadha
force-pushed
the
pchadha/async-vllm
branch
from
09c45a8 to
fb47ba9
Compare
parthchadha
added
CI:L0
terrykong
reviewed
May 20, 2025
There was a problem hiding this comment.
do you mind adding this key to all the grpo recipes under examples/config/recipes/grpo*.yaml?
terrykong
reviewed
May 20, 2025
| # Reset the prefix cache to ensure that prefix cache is not reused after weights are updated | ||
| self.llm.llm_engine.reset_prefix_cache() | ||
| self.llm.sleep(level=1) | ||
| async def sleep(self): |
There was a problem hiding this comment.
i don't see a synchronous analog to some of these async functions. Is the expectation for users using LLMEngine to still await or should we provide a synchronous variant?
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Outdated suggestions cannot be applied.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.
What does this PR do ?
This PR adds the capability to use async vllm engine and verify its correctness in unit tests. This is first of N PR's that will enable async processing of rollout requests.
Issues
List issues that this PR closes (syntax):
Usage
# Add a code snippet demonstrating how to use thisBefore your PR is "Ready for review"
Pre checks:
Additional Information