Create vLLM v1 benchmark dashboard #6306
Merged
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
|
@huydhn is attempting to deploy a commit to the Meta Open Source Team on Vercel. A member of the Team first needs to authorize it. |
facebook-github-bot
added
the
CLA Signed
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
huydhn
commented
Feb 21, 2025
| : `${model} (${dtype} / ${device})`; | ||
| if (repoName === "vllm-project/vllm") { | ||
| let requestRate = record.extra!["request_rate"]; | ||
| // TODO (huydhn): Fix the invalid JSON on vLLM side |
There was a problem hiding this comment.
Here is the fix on vLLM side vllm-project/vllm#13641 under review. Once this lands, we can remove these hacks
clee2000
approved these changes
Feb 21, 2025
Co-authored-by: clee2000 <[email protected]>
yangw-dev
reviewed
Feb 21, 2025
| </a> | ||
| . {children} | ||
| </Typography> | ||
| {repoName !== "vllm-project/vllm" && ( |
There was a problem hiding this comment.
option: maybe can place a list of repoName that has this exception
There was a problem hiding this comment.
Yeah, the benchmark UX seems to be growing pretty fast recently, so I think I will take a step back to see if we could refactor the code here for better modularization
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Outdated suggestions cannot be applied.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.
This is the initial version of vLLM v1 benchmark dashboard. The benchmark is run periodically on vLLM main commits. The script running the benchmark is at https://github.com/pytorch/pytorch-integration-testing/tree/master/vllm-benchmarks.
Besides all the custom logic for
vllm-project/vllm, I also add a newextramap in the query to store arbitrary information about how the benchmark is setup.Some UX features are left for subsequent PRs:
request_rateandtensor_parallel_sizeare missing when the former is set toInfleading to an invalid JSON. This fix needs to be done on vLLM sidePreview
https://torchci-git-fork-huydhn-create-vllm-benchma-323632-fbopensource.vercel.app/benchmark/llms?repoName=vllm-project%2Fvllm