Skip to content

[Superseded] Kimi-K2.5 MI355X tuning draft / [已替代] Kimi-K2.5 MI355X 调优草稿#2212

Closed
Fangzhou-Ai wants to merge 1 commit into
mainfrom
perf/kimi-k2.5-mi355x-a10
Closed

[Superseded] Kimi-K2.5 MI355X tuning draft / [已替代] Kimi-K2.5 MI355X 调优草稿#2212
Fangzhou-Ai wants to merge 1 commit into
mainfrom
perf/kimi-k2.5-mi355x-a10

Conversation

@Fangzhou-Ai

@Fangzhou-Ai Fangzhou-Ai commented Jul 14, 2026

Copy link
Copy Markdown
Collaborator

Superseded

This draft was replaced by #2213 after cleaning the branch and PR metadata. Please review the active draft instead; no changes from this closed PR should be merged.

已替代

清理分支和 PR 元数据后,本草稿已由 #2213 替代。请改为审核新的草稿 PR;本已关闭 PR 中的改动不应合并。

@github-actions

Copy link
Copy Markdown
Contributor

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

@Fangzhou-Ai Fangzhou-Ai force-pushed the perf/kimi-k2.5-mi355x-a10 branch from 4469d9d to 1117386 Compare July 14, 2026 22:32
@Fangzhou-Ai

Copy link
Copy Markdown
Collaborator Author

This PR can only be enabled to merge after we bump the aiter version for vLLM.

Comment thread configs/amd-master.yaml
kimik2.5-fp4-mi355x-vllm:
image: vllm/vllm-openai-rocm:v0.24.0
# TODO: replace with the official vLLM ROCm image containing AITER #3832.
image: TBD

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@chunfangamd after this place holder is filled, then we can start the perf sweep.

Apply the accuracy-gated TP4 settings, retain the full concurrency sweep, and keep the vLLM ROCm image as a draft placeholder pending the required AITER bump.

中文:应用已通过精度验证的 TP4 配置,保留完整并发扫描范围,并暂时保留 vLLM ROCm 镜像占位符,等待所需的 AITER 版本升级。
@Fangzhou-Ai Fangzhou-Ai force-pushed the perf/kimi-k2.5-mi355x-a10 branch from 1117386 to bc9fa8a Compare July 14, 2026 22:45
@Fangzhou-Ai Fangzhou-Ai deleted the perf/kimi-k2.5-mi355x-a10 branch July 14, 2026 22:45
@Fangzhou-Ai Fangzhou-Ai changed the title [AMD] Tune Kimi-K2.5 MXFP4 MI355X vLLM toward ATOM parity / 调优 Kimi-K2.5 MXFP4 MI355X vLLM,缩小与 ATOM 的性能差距 [Superseded] Kimi-K2.5 MI355X tuning draft / [已替代] Kimi-K2.5 MI355X 调优草稿 Jul 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Development

Successfully merging this pull request may close these issues.

1 participant