[Klaud Cold] Update minimaxm3-fp8-mi300x-vllm-agentic-mtp vLLM ROCm image to v0.29.0 / 将 minimaxm3-fp8-mi300x-vllm-agentic-mtp 的 vLLM ROCm 镜像更新至 v0.29.0 - #3063
Conversation
… v0.29.0 Move the MI300X MiniMax-M3 MXFP8 AgentX EAGLE3 recipe from the unstable commit nightly vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 to the v0.29.0 release image vllm/vllm-openai-rocm:v0.29.0 (digest sha256:e5e47f6aaab675c252c381f0dac237b31b10d87bb74d092b07fb4065efd7f5a1). Recipe, search space and evals are unchanged. 将 MI300X MiniMax-M3 MXFP8 AgentX EAGLE3 配方的镜像从不稳定的提交 nightly vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 更新为 v0.29.0 发布镜像 vllm/vllm-openai-rocm:v0.29.0 （digest sha256:e5e47f6aaab675c252c381f0dac237b31b10d87bb74d092b07fb4065efd7f5a1）。 配方、搜索空间与评测保持不变。 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
|
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 For PR verification, add the 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 如需进行 PR 验证，请为此 PR 添加 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 |
|
Baseline: 2026-09-08 · AgentX · TP8/EP1 · Mean latency
中文**基线：**2026-09-08 · |
|
Initial attempt · Passed · Run 34729027628 / attempt 1 · 2026-09-13 02:35 UTC
Note: All rows: Δ N/A: no matched eval baseline. Next: Append the perf-changelog entry, validate the full matrix and start the final full sweep with full-sweep-enabled. 中文初次尝试 · 已通过 · Run 34729027628 / attempt 1 · 2026-09-13 02:35 UTC |
…8ec8ca-4f7436dbcc1df172
…tic-mtp v0.29.0 image bump Record the vLLM ROCm image move from nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 to the v0.29.0 release image for PR #3063 after the smoke run passed. 为 PR #3063 记录 vLLM ROCm 镜像从 nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36 更新至 v0.29.0 发布镜像的 perf-changelog 条目（冒烟运行已通过）。 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
|
Final full sweep · Passed · Run 34733463164 / attempt 1 · 2026-09-13 05:14 UTC
Note: c16: 2 request errors.
Note: All rows: Δ N/A: no matched eval baseline. Next: Mark ready for maintainer review. 中文最终完整 sweep · 已通过 · Run 34733463164 / attempt 1 · 2026-09-13 05:14 UTC |
|
see unofficial run visualizer at https://inferencex.semianalysis.com/inference?unofficialRun=34733463164 |
|
validated · Repairs: 0 · Runs: 34729027628, 34733460912, 34733463164 中文validated · 修复次数：0 · 运行：34729027628, 34733460912, 34733463164 |
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
There was a problem hiding this comment.
Looks good, straightforward image-bump config change.
What was reviewed: the image field bump in configs/amd-master.yaml for minimaxm3-fp8-mi300x-vllm-agentic-mtp (nightly tag to v0.29.0 release tag); confirmed the matching perf-changelog.yaml entry is appended at the true tail without disturbing prior bytes, and that it follows the existing config-keys/description/pr-link schema with paired English/Chinese descriptions like neighboring entries. This recipe is single-node (multinode: false), so the multi-node model.container == image consistency rule doesn't apply here.
Extended reasoning...
Overview
The diff touches exactly two files: configs/amd-master.yaml (a one-line image: bump from a ROCm nightly tag to the v0.29.0 release tag for the minimaxm3-fp8-mi300x-vllm-agentic-mtp recipe) and perf-changelog.yaml (a new append-only entry documenting that same change, with English and Chinese description bullets, config-keys, and a pr-link). No benchmark scripts, search-space definitions, evals, or other recipes were modified.
Security risks
None. This is a config data change pointing to an upstream, tagged Docker Hub image (vllm/vllm-openai-rocm:v0.29.0), not a build script, credential, or executable code path. No injection, auth, or data-exposure surface is touched.
Level of scrutiny
This warrants light scrutiny consistent with a mechanical version-bump PR. I verified the changelog entry lands strictly after the prior entry at line 7443 (no reordering or deletion of existing content) and that its shape (config-keys / description list with EN+ZH bullets / pr-link) mirrors the immediately preceding entries in the file. The multinode: false setting on this recipe means the "model.container must equal image" multi-node rule cited in repo conventions doesn't apply, so there's no cross-field consistency issue to flag.
Other factors
No bug-hunting findings were reported, and my own reading of the diff didn't surface anything beyond what's already documented in the changelog entry (release-branch exclusions of two upstream vLLM PRs, noted as informational rather than a functional gap in this change). The change is small, self-contained, and follows an established repository pattern for image-bump PRs.
1 participant
Goal: Update vLLM image from
vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36tovllm/vllm-openai-rocm:v0.29.0.Baseline: 2026-09-08 ·
vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36AgentX · TP8/EP1 · Mean latency · Sources: API 1, API 2, API 3, API 4, API 5
中文
**目标：**将 vLLM 镜像从
vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36更新为vllm/vllm-openai-rocm:v0.29.0。**基线：**2026-09-08 ·
vllm/vllm-openai-rocm:nightly-d9105ea8001e0a6d77a96327d17515bb5791fb36AgentX · TP8/EP1 · 平均延迟 · 来源： API 1, API 2, API 3, API 4, API 5；数值及异常说明见上表。