Update minimaxm2.5-fp4-mi355x-vllm vLLM image to v0.20.2#1319
Update minimaxm2.5-fp4-mi355x-vllm vLLM image to v0.20.2#1319functionstackx wants to merge 1 commit into
Conversation
Co-authored-by: Klaud Cold <Klaud-Cold@users.noreply.github.com>
|
Thanks for the contribution! For vLLM & SGLang, please ensure that your recipes is similar to the official vLLM recipes and/or the SGLang cookbook If it is not, please create a PR first before we can merge your single node PR into the master branch. Let's ensure that the documentation is first class such that the entire ML community can benefit from your hard work! Thank you PR authors are responsible for ensuring that after merging, all GitHub Action jobs fully pass. A lot of the time, failures are just flakes and simply re-running the failed jobs will fix it. If re-running failed jobs is attempted, PR authors are responsible for ensuring it passes. See GitHub's docs on re-running failed jobs: https://docs.github.com/en/actions/how-tos/manage-workflow-runs/re-run-workflows-and-jobs#re-running-failed-jobs-in-a-workflow As a rule of thumb, generally, PR authors should request a review & get a PR approval from the respective companies' CODEOWNERS before requesting a review from core maintainers. If additional help is needed, PR authors can reach out to core maintainers over Slack. |
| - minimaxm2.5-fp4-mi355x-vllm | ||
| description: | ||
| - "Update vLLM image from v0.19.1 to v0.20.2" | ||
| pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX |
There was a problem hiding this comment.
🟡 The new perf-changelog.yaml entry uses pull/XXX as a placeholder instead of the actual PR number. Since this is PR #1319, the link should be https://github.com/SemiAnalysisAI/InferenceX/pull/1319; as-written it 404s and the changelog row is non-navigable.
Extended reasoning...
What the bug is
The new entry added to perf-changelog.yaml at line 2351 contains a placeholder XXX in its pr-link value:
- config-keys:
- minimaxm2.5-fp4-mi355x-vllm
description:
- "Update vLLM image from v0.19.1 to v0.20.2"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXXThis appears to be a template value that was never filled in before submission.
Why this is wrong
Every other entry in perf-changelog.yaml follows the convention of using a real, numeric PR link. The five entries immediately preceding this one all reference real PR numbers (e.g. #1303, #1304, #1305, #1308, #1310). The PR being submitted here is #1319 (per the PR metadata), so the link should read:
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1319Impact
As-is, the URL https://github.com/SemiAnalysisAI/InferenceX/pull/XXX produces a GitHub 404 when followed, so the new changelog row is non-navigable. Anyone consuming this changelog (humans, dashboards, or downstream automation that resolves these links) won't be able to trace this perf-changelog entry back to the PR that introduced the change. It also breaks the file-wide convention that lets readers cross-reference perf changes with their originating PRs.
Step-by-step proof
- The PR metadata above reports
<pr number="1319">. - The diff for
perf-changelog.yamladds exactly one entry whosepr-linkishttps://github.com/SemiAnalysisAI/InferenceX/pull/XXX. - Resolving that URL: GitHub has no pull request with id
XXX(non-numeric), so any client following the link gets a 404. - Substituting the real PR number gives
https://github.com/SemiAnalysisAI/InferenceX/pull/1319, which is the canonical link for this PR and matches the format used by all prior entries in the file.
How to fix
Replace pull/XXX with pull/1319 on line 2351 of perf-changelog.yaml. No other changes needed.
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25761613027 |
Summary
minimaxm2.5-fp4-mi355x-vllmimage fromvllm/vllm-openai-rocm:v0.19.1tovllm/vllm-openai-rocm:v0.20.2Ref #1154
Generated with Claude Code