Skip to content

vllm-dp-support #3303

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

vllm-dp-support #3303

wants to merge 1 commit into from

Conversation

shirinyamani
Copy link
Member

What does this PR do?

In this PR we create multiple LLM instances (one per DP rank) and distribute the prompts among them.
From data_parallel.py, we can see how to set the environment variables for DP communication and how to split the prompts among DP ranks.
So we do;

  1. Initialize environment variables for DP
  2. Create multiple LLM instances, one for each DP rank
  3. Modify the generate to distribute prompts across DP ranks and collect results

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you read the contributor guideline,
    Pull Request section?
  • Was this discussed/approved via a GitHub issue? Please add a link
    to it if that's the case.
  • Did you make sure to update the documentation with your changes?
  • Did you write any new necessary tests?

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.

@HuggingFaceDocBuilderDev

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.

@shirinyamani shirinyamani requested a review from Copilot April 16, 2025 06:55
Copy link

@Copilot Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot reviewed 1 out of 1 changed files in this pull request and generated no comments.

Comment on lines +438 to +442
for dp_rank, batch in enumerate(prompt_batches):
if batch: # Skip empty batches
llm_instance = llm_instances[dp_rank]
batch_outputs = llm_instance.generate(batch, sampling_params=sampling_params)
all_outputs.extend(batch_outputs)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is it sequential?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants