Update kimik2.5-fp4-b200-vllm vLLM image to v0.20.2#1336
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. |
1 similar comment
|
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. |
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763438871 |
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
| - kimik2.5-fp4-b200-vllm | ||
| description: | ||
| - "Update vLLM image from v0.17.0 to v0.20.2" | ||
| pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX |
There was a problem hiding this comment.
🟡 The new perf-changelog entry uses the literal placeholder pull/XXX instead of the actual PR number (1336), so the link will 404. Replace /pull/XXX with /pull/1336 on line 2351 to match the convention used by every other entry in the file.
Extended reasoning...
What's wrong
The diff adds a new perf-changelog entry whose pr-link is:
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXXThe trailing XXX is a literal placeholder string that was clearly meant to be substituted with the actual PR number before the PR was opened, but was left unreplaced.
Why this is a real defect
Every other entry in perf-changelog.yaml uses a concrete PR number — for example, the five entries immediately preceding this one use /pull/1303, /pull/1304, /pull/1305, /pull/1308, and /pull/1310. The convention is consistent across the entire file. /pull/XXX is not a valid GitHub path: visiting https://github.com/SemiAnalysisAI/InferenceX/pull/XXX returns a 404, so any reader (human or downstream tooling) following the link from this changelog entry will get a broken page instead of the merge that introduced the change.
Step-by-step proof
- Open
perf-changelog.yamlat the new entry (lines 2347–2351 after merge). - The last line reads
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX. - Compare with line 2345 immediately above:
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1310— a numeric PR id. - GitHub PR paths are of the form
/pull/<integer>;/pull/XXXdoes not match an existing PR and resolves to a 404 page. - This PR is Update kimik2.5-fp4-b200-vllm vLLM image to v0.20.2 #1336 (see the PR metadata), so the intended value is
1336.
Fix
Change line 2351 from:
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXXto:
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1336This is a one-character-class, mechanical edit. The change is cosmetic in terms of runtime behavior (the changelog itself doesn't affect benchmarks or images) but breaks the file's link convention and any tooling that parses pr-link and follows it, so it's worth correcting before merge.
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763439479 |
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25818238248 |
3 similar comments
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25818238248 |
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25818238248 |
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25818238248 |
2 participants
Summary
kimik2.5-fp4-b200-vllmimage fromvllm/vllm-openai:v0.17.0tovllm/vllm-openai:v0.20.2Ref #1154
Generated with Claude Code