Update gptoss-fp4-mi300x-vllm vLLM image to v0.20.2

[Klaud-Cold]

Summary

• Update gptoss-fp4-mi300x-vllm image from vllm/vllm-openai-rocm:v0.17.0 to vllm/vllm-openai-rocm:v0.20.2

Ref #1154

Generated with Claude Code

[github-actions]

Thanks for the contribution! For vLLM & SGLang, please ensure that your recipes is similar to the official vLLM recipes and/or the SGLang cookbook

• https://docs.sglang.io/cookbook/intro
• https://recipes.vllm.ai/

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

• https://github.com/vllm-project/recipes
• https://github.com/sgl-project/sglang/tree/main/docs_new

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.

[claude]

🔴 The new perf-changelog entry for this image bump uses a placeholder pr-link https://github.com/SemiAnalysisAI/InferenceX/pull/XXX instead of the actual PR number. This produces a 404 link in the rendered changelog and breaks any tooling that parses pr-link to associate the entry with a real PR. Please replace XXX with 1349 (or the final PR number) before merging.

Extended reasoning...

What the bug is

In perf-changelog.yaml at lines 2347-2351, the new changelog entry added by this PR ends with:

- config-keys:
    - gptoss-fp4-mi300x-vllm
  description:
    - "Update vLLM ROCm image from v0.17.0 to v0.20.2"
  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX

The literal string XXX was never substituted with the actual PR number.

Why this is wrong

Every other entry in the file uses a real numeric PR ID. The immediately preceding entries reference PRs #1308 (line 2338) and #1310 (line 2345), and entries further up reference #1303, #1304, #1305, etc. The established convention — used consistently across the entire file — is that pr-link points at the merging PR with a real numeric ID.

Step-by-step proof

1. The PR metadata above identifies this as PR Update gptoss-fp4-mi300x-vllm vLLM image to v0.20.2 #1349.
2. The diff adds the following block to perf-changelog.yaml:

   - config-keys:
       - gptoss-fp4-mi300x-vllm
     description:
       - "Update vLLM ROCm image from v0.17.0 to v0.20.2"
     pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX

3. Visiting https://github.com/SemiAnalysisAI/InferenceX/pull/XXX resolves to a 404 because XXX is not a valid GitHub PR number.
4. Any consumer of perf-changelog.yaml that parses pr-link (e.g. to deep-link entries to their PR, or to cross-reference commits to PRs) will either fail to parse the URL as an integer or produce a dangling link.

Note on the referenced issue

The PR description references #1154 (the gptoss tracking issue). That reference belongs in the body of the PR, not in pr-link — pr-link in perf-changelog.yaml consistently points at the merging PR, not at related issues. The correct value here is https://github.com/SemiAnalysisAI/InferenceX/pull/1349.

How to fix

Replace XXX with 1349:

  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1349

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763470990
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=25763470990

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763470990
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=25763470990

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763470990
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=25763470990