[Klaud Cold] Update kimik2.5-int4-mi355x-vllm vLLM ROCm image to v0.22.0

[functionstackx]

Summary

Update vLLM ROCm image from v0.21.0 to v0.22.0

Recipes touched: kimik2.5-int4-mi355x-vllm

Test plan

• full-sweep-enabled sweep passes.

🤖 Generated with Claude Code

---

Note

Low Risk
Single benchmark image tag change with no serving-flag or sweep-matrix edits; main risk is v0.22.0 runtime regressions on MI355X, expected to be caught by the planned full sweep.

Overview
Bumps the Kimi-K2.5 INT4 MI355X vLLM benchmark container from vllm/vllm-openai-rocm:v0.21.0 to v0.22.0 in .github/configs/amd-master.yaml for config key kimik2.5-int4-mi355x-vllm. Model, runner, TP=8, and fixed-seq-len sweeps are unchanged.

Adds a matching perf-changelog.yaml entry so the image bump is tracked for perf history.

^{Reviewed by Cursor Bugbot for commit 4141462. 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=26696244526
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26696244526

[claude]

LGTM, straightforward image version bump.

Extended reasoning...

Overview

This PR bumps the vLLM ROCm image for the kimik2.5-int4-mi355x-vllm recipe from v0.21.0 to v0.22.0 in .github/configs/amd-master.yaml, with a matching entry added to perf-changelog.yaml. Only one recipe is touched; all other recipes are unchanged.

Security risks

None. This is a config-only change selecting a different published vLLM ROCm image tag. No code, auth, crypto, or permission surfaces are touched.

Level of scrutiny

Minimal. This is exactly the kind of mechanical config bump the approval criteria call out — single-line image tag change plus a corresponding changelog entry. The PR carries the full-sweep-enabled label, so the actual functional validation happens via the sweep CI, not via human code review.

Other factors

No bugs reported by the bug hunting system, no prior reviewer comments outstanding, and the changelog format mirrors recent precedent in the file.

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=26696245954
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=26696245954