[Klaud Cold] Update minimaxm3-fp8-h200-vllm-agentic-mtp vLLM image to nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36

[functionstackx]

Summary

Update vLLM image for the H200 MiniMax-M3 MXFP8 AgentX EAGLE3 recipe from vllm/vllm-openai:v0.27.1 (v0.27.1 release) to vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 (2026-09-07 upstream nightly).

• Tag verified on Docker Hub: last pushed 2026-09-07T06:16:01Z, digest sha256:254eebf919e8b7b0d530d97fccc606c36f380ff6724d64bc190951bec1aee838; tag commit vllm-project/vllm@d9105ea8, the same tag the B200 MiniMax-M3 AgentX recipe moved to in [Klaud Cold] Update minimaxm3-fp4-b200-vllm-agentic-mtp vLLM image to nightly-8a728663c1c3eeace834a95f5654fa653cc1998c #2860.
• benchmarks/single_node/agentic/minimaxm3_fp8_h200_mtp.sh is unchanged and pins nothing image-specific: TRITON_ATTN attention, fp8 KV, EAGLE3 with the Inferact EAGLE3-GQA draft on FLASH_ATTN, golden synthetic AL 2.78, Mooncake 0.3.11.post1 on the DRAM-offload arm. Grid unchanged: resident TP8 c1/c2/c4/c6/c8/c10 and Mooncake DRAM offload c12/c14.

Recipes touched: minimaxm3-fp8-h200-vllm-agentic-mtp

Test plan

• full-sweep-enabled sweep passes on cluster:h200-dgxc.

🤖 Generated with Claude Code

---

Note

Low Risk
Only swaps a pinned container image and documents it in the perf changelog; no script or search-space changes.

Overview
Updates the minimaxm3-fp8-h200-vllm-agentic-mtp recipe to use vllm/vllm-openai:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 instead of vllm/vllm-openai:v0.27.1, aligning the H200 AgentX EAGLE3 run with the same upstream nightly already used on B200 (#2860) and related ROCm bumps.

Adds a perf-changelog.yaml entry for the agentic-coding scenario. Benchmark script, Mooncake offload settings, and the TP8 concurrency grid are unchanged.

^{Reviewed by Cursor Bugbot for commit 8bd1898. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Thanks for the contribution! Please reach out to respective companies' CODEOWNER to fill in the latest PR_REVIEW_CHECKLIST.md before pinging core maintainer on Slack for review. In order for the signoff PR check bot to trigger, you must follow the PR_REVIEW_CHECKLIST.md template correctly, including the phrase As a PR reviewer and CODEOWNER, I have reviewed this and have.

For PR verification, add the full-sweep-fail-fast label (strongly recommended) to this PR — the benchmark sweep only runs on labeled PRs. Use full-sweep-enabled only if you need matrix jobs to keep running past a failure.

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. See GitHub's docs on re-running failed jobs

---

感谢你的贡献！请联系相应公司的 CODEOWNER 填写最新的 PR_REVIEW_CHECKLIST.md，然后再在 Slack 上联系核心维护者进行审阅。为了触发 signoff PR 检查机器人，你必须正确遵循 PR_REVIEW_CHECKLIST.md 模板，包括保留英文语句 As a PR reviewer and CODEOWNER, I have reviewed this and have。

如需进行 PR 验证，请为此 PR 添加 full-sweep-fail-fast 标签（强烈推荐）— 基准测试 sweep 仅在带有标签的 PR 上运行。仅当需要矩阵任务在失败后继续运行时才使用 full-sweep-enabled。

PR 作者有责任确保合并后所有 GitHub Action 任务完全通过。 很多时候失败只是偶发抖动（flake），重新运行失败的任务即可解决。参见 GitHub 关于重新运行失败任务的文档

[claude]

🔴 The new entry's pr-link: PRLINK_PLACEHOLDER is not a recognized value, so utils/validate_perf_changelog.py's validate_added_pr_link will reject this PR's changelog gate check (it only accepts the real pull/<N> link or the literal XXX/.../pull/XXX placeholders). Fix: use pr-link: XXX (or the canonical https://github.com/SemiAnalysisAI/InferenceX/pull/XXX) as the placeholder so CI passes until merge_with_reuse.sh fills in the real link, per KLAUD_DEBUG.md's canonical append pattern.

Extended reasoning...

validate_perf_changelog.py:144-160 validate_added_pr_link() checks the appended entry's pr-link: if a pr_number is known it must equal the expected .../pull/<N> string or be one of PR_LINK_PLACEHOLDERS = {"XXX", ".../pull/XXX"} (validate_perf_changelog.py:24-27). PRLINK_PLACEHOLDER matches neither, so validate_added_pr_link raises ChangelogValidationError and the changelog gate CI check fails on this PR, blocking merge until the entry is corrected to a valid placeholder or the real link.

Verification: normal — but with a mechanism correction: the candidate's root cause is right, its cited trigger point is wrong. perf-changelog.yaml:6931 commits pr-link: PRLINK_PLACEHOLDER, which is neither the expected .../pull/<N> URL nor a member of PR_LINK_PLACEHOLDERS = {"XXX", ".../pull/XXX"} (validate_perf_changelog.py:24-27), so validate_added_pr_link (144-160) raises… | normal (with a…

[github-actions]

see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=34174440759
see unofficial run visualizer at https://inferencex.semianalysis.com/evaluation?unofficialRun=34174440759

[functionstackx]

/reuse-sweep-run