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.

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=26902572165
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26902572165

Claude finished @jasonlizhengjian's task in 1m 49s —— View job

Review of PR #1658

• Read changed files and PR diff
• Validate master config changes
• Validate perf-changelog.yaml
• Validate benchmark scripts (EP pattern, code style)
• Post review findings

LGTM — no blocking issues found.

What was checked:

• Master config + perf-changelog: nvidia-master.yaml modified and perf-changelog.yaml updated with a new entry appended to the end of the file. ✓
• Container images: vllm/vllm-openai:v0.22.1 is publicly accessible on Docker Hub. ✓
• EP conditional pattern: Both B200 and B300 scripts use the correct if [ "$EP_SIZE" -gt 1 ] guard with --enable-expert-parallel (boolean flag, as required for vLLM). ✓
• check_env_vars: Both scripts now require EP_SIZE. ✓
• Code style: vllm serve arguments are on separate lines with \ continuations. ✓
• Sweep points: EP points are symmetric across B200/B300 and across both ISL scenarios (1k/1k and 8k/1k). ✓

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27227919547
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27227919547

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=27228874121
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=27228874121