[Klaud Cold] Update kimik2.5-int4-b300-vllm vLLM image to v0.22.0

[functionstackx]

Summary

Update vLLM image from v0.21.0 to v0.22.0

Recipes touched: kimik2.5-int4-b300-vllm

Test plan

• full-sweep-enabled sweep passes.

🤖 Generated with Claude Code

---

Note

Low Risk
Single image pin for an existing benchmark config with no serve flags or search-space changes; impact is limited to sweep/runtime compatibility with the newer vLLM build.

Overview
Bumps the Kimi-K2.5 INT4 B300 vLLM benchmark container from vllm/vllm-openai:v0.21.0 to v0.22.0 in nvidia-master.yaml. Model, runner, search space, and the note that B300 still follows the B200 recipe are unchanged.

Adds a matching perf-changelog.yaml entry for config key kimik2.5-int4-b300-vllm.

^{Reviewed by Cursor Bugbot for commit 1483da6. Bugbot is set up for automated code reviews on this repo. Configure here.}

[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]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=26696216144
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26696216144

[claude]

LGTM — single-recipe vLLM image tag bump (v0.21.0 → v0.22.0) with a matching perf-changelog entry.

Extended reasoning...

Overview

This PR bumps the Docker image tag for the kimik2.5-int4-b300-vllm recipe in .github/configs/nvidia-master.yaml from vllm/vllm-openai:v0.21.0 to v0.22.0, and adds a matching entry in perf-changelog.yaml to track the bump. No other fields in the recipe (model, runner, scenarios, search space, serve flags) are changed.

Security risks

None. This is a benchmark/CI config change that updates an external image tag. The image is from the same upstream vllm/vllm-openai repository already in use across many other recipes in this file.

Level of scrutiny

Low. This is a mechanical version bump touching a single recipe and a changelog. It follows the established pattern used throughout nvidia-master.yaml and perf-changelog.yaml. The full-sweep-enabled label indicates the perf sweep will validate the change before merge.

Other factors

Bug hunting system found no issues. Cursor Bugbot also flagged this as low risk. The changelog entry is correctly formatted and matches the convention of surrounding entries.

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=26696217452
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26696217452