Update qwen3.5-fp4-b300-sglang SGLang image to v0.5.11-cu130

[Klaud-Cold]

Summary

• Update qwen3.5-fp4-b300-sglang image from lmsysorg/sglang:v0.5.10.post1-cu130 to lmsysorg/sglang:v0.5.11-cu130

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.

[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.

[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 has pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX — the literal placeholder XXX was never replaced with the actual PR number (1343). All other entries in the file use real numeric PR numbers (e.g., /pull/1304, /pull/1305, /pull/1308, /pull/1310). Fix by replacing XXX with 1343.

Extended reasoning...

What the bug is

The PR appends a new entry to perf-changelog.yaml documenting the SGLang image bump for qwen3.5-fp4-b300-sglang. The added entry ends with:

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

The XXX is a literal placeholder string that should have been substituted with the real PR number — 1343 — before opening the PR.

Why this matters / how it manifests

The pr-link field is the changelog's authoritative reference back to the PR that introduced a given change. Every other entry in perf-changelog.yaml follows the convention of including a real numeric PR ID (visible in the surrounding context: /pull/1304, /pull/1305, /pull/1308, /pull/1310). With XXX in place, the URL https://github.com/SemiAnalysisAI/InferenceX/pull/XXX resolves to a 404 — the changelog reference is broken and useless for anyone trying to follow the link to context, discussion, or review history.

Why existing safeguards don't prevent it

There is no schema validation or CI check that enforces a numeric value in the pr-link field, so a placeholder slips through unnoticed. The PR is also marked full-sweep-enabled, suggesting it is intended to be merged once benchmarks pass — there is no human guardrail catching this string before merge.

Step-by-step proof

1. Open the PR diff. The final hunk in perf-changelog.yaml adds (literally, with the XXX):

   - config-keys:
       - qwen3.5-fp4-b300-sglang
     description:
       - "Update SGLang image from v0.5.10.post1-cu130 to v0.5.11-cu130"
     pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX

2. Compare against the immediately preceding entry (also in the diff context), which uses the real value https://github.com/SemiAnalysisAI/InferenceX/pull/1310.
3. Navigate to https://github.com/SemiAnalysisAI/InferenceX/pull/XXX — GitHub returns a 404 because XXX is not a valid PR number.
4. The actual PR is Update qwen3.5-fp4-b300-sglang SGLang image to v0.5.11-cu130 #1343 (per the PR metadata), so the intended URL is https://github.com/SemiAnalysisAI/InferenceX/pull/1343.

How to fix

Replace XXX with 1343 on line 2351 of perf-changelog.yaml:

-  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1343

This is a trivial documentation-only fix but should be done before merge so the changelog entry is actually useful to future readers.

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25763457522
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=25763457522

[Oseltamivir]

/reuse-sweep-run

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=25783450713
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=25783450713