-
Notifications
You must be signed in to change notification settings - Fork 36
feat: Add MoE support for VLLM #118
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
ilyasher
merged 4 commits into
dev-isherstyuk-refactor-moe-helper-funcs
from
dev-isherstyuk-vllm-moe
Nov 26, 2025
Merged
feat: Add MoE support for VLLM #118
ilyasher
merged 4 commits into
dev-isherstyuk-refactor-moe-helper-funcs
from
dev-isherstyuk-vllm-moe
Nov 26, 2025
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
5921c78 to
b7db778
Compare
|
tianhaox
reviewed
Nov 24, 2025
Contributor
|
Signed-off-by: Ilya Sherstyuk <[email protected]>
Signed-off-by: Ilya Sherstyuk <[email protected]>
Signed-off-by: Ilya Sherstyuk <[email protected]>
ff4e819 to
2f5b658
Compare
Signed-off-by: Ilya Sherstyuk <[email protected]>
67ea728 to
c209445
Compare
Contributor
Author
|
I found my mistake when ep>1, so I updated the collector and re-collected the h100 moe perf. I also updated enumerate_parallel_config() as suggested by @tianhaox |
b17ba13
into
dev-isherstyuk-refactor-moe-helper-funcs
6 of 7 checks passed
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Outdated suggestions cannot be applied.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.
Add MoE collector for VLLM and collect data for H100
Perf results
I only collected h100 data for now. It is poor compared to TRTLLM. While TRTLLM is about ~30% of calculated SOL, the collected VLLM data is <10% of SOL. I suspect this is because the TRTLLM data collector ran autotune for MoE kernels, but vllm did not. vllm does have a autotuner for MoE, but it takes hours per test case, so it is not feasible to run it for our data collection.
VLLM MoE sanity check

TRTLLM for reference
