Skip to content

[ROCm][MoE configs] mi325 mixtral & mi300 qwen_moe #13503

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

Merged
merged 2 commits into from
Feb 19, 2025

Conversation

divakar-amd
Copy link
Contributor

@divakar-amd divakar-amd commented Feb 18, 2025

This PR add the following moe configs:

  • MI325 configs Mixtral-8x(7B,22B) TP=(1,2,4,8) FP=(8,16)
  • MI300 configs qwen1.5_moe_A2.7B TP=(1,2,4,8) fp16

Signed-off-by: Divakar Verma <divakar.verma@amd.com>
Signed-off-by: Divakar Verma <divakar.verma@amd.com>
Copy link

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

@robertgshaw2-redhat
Copy link
Collaborator

Im assuming you have run a perf benchmark on this.

Also, don't we need OAM versions of these too?

@robertgshaw2-redhat robertgshaw2-redhat added the ready ONLY add when PR is ready to merge/full CI is needed label Feb 19, 2025
@robertgshaw2-redhat robertgshaw2-redhat enabled auto-merge (squash) February 19, 2025 03:13
@simon-mo simon-mo merged commit 8aada19 into vllm-project:main Feb 19, 2025
58 of 60 checks passed
@divakar-amd
Copy link
Contributor Author

@robertgshaw2-redhat We won't need OAM versions going forward. The following PR will get the correct (& unique) names for AMD GPUs without any trailing "_OAM": #13438

xjpang pushed a commit to xjpang/vllm that referenced this pull request Feb 20, 2025
lulmer pushed a commit to lulmer/vllm that referenced this pull request Apr 7, 2025
Signed-off-by: Louis Ulmer <ulmerlouis@gmail.com>
shreyankg pushed a commit to shreyankg/vllm that referenced this pull request May 3, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ready ONLY add when PR is ready to merge/full CI is needed
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants