Skip to content

[VideoGenX] run matched Wan workloads on SGLang and vLLM-Omni - #3733

Closed
edwingao28 wants to merge 60 commits into
mainfrom
feat/wan-serving-runner
Closed

edwingao28 wants to merge 60 commits into
mainfrom
feat/wan-serving-runner

Conversation

@edwingao28

@edwingao28 edwingao28 commented Oct 5, 2026 •

Copy link
Copy Markdown
Collaborator

Description

Run Wan2.2 on SGLang and vLLM-Omni through native sweeps with a shared 20-clip workload, immutable plans, and sealed failure evidence.

Validation Result
Producer → App Both synthetic engine bundles passed archive-to-History regression.
CPU CI Tests/lint/harness/spec/power passed at 4475151c.
H100/a8576ccb One clip, submission through validation: 37.97 s; 4518 W across eight GPUs; informational only.

Blocker: Depends on #4243. Native H200 GPU qualification remains unverified.

Related Issue

#2916, #4243, #4246.

Type of Change and Checklist

Type of Change

  • Bug fix
  • New feature
  • Configuration change
  • Documentation update
  • Other (please describe)

Checklist

  • I have tested my changes locally
  • I have updated documentation if necessary
  • For every change that can affect benchmark performance and every recipe addition or modification, I have appended a new entry to the physical end of inferencex-e2e/perf-changelog.yaml and have not edited historical entries
  • Before merging via reuse, an authorized maintainer (OWNER/MEMBER/COLLABORATOR) has commented /use <run_id> (or the legacy /reuse-sweep-run) on this PR. Do this only once there is a final full sweep that is all green with evals passing, since after this comment the sweep label will no longer automatically kick off new sweeps. Remove and re-add the label to force one.

edwingao28 and others added 7 commits September 26, 2026 23:23
移植既有 H3 视频流程到当前主线,保留独立资源生命周期和结果契约;对齐正式 changelog 解析、暂存功耗模块、AMD SMI 与封装脚本校验。
补回 H3 父 PR 已有的测试清理:删除仅回显 stub 输入的断言,保留 CLI 退出行为验证。
在当前主线移除 utils 兼容入口后,改用 infx 功耗模块与优先级 CLI,保留隔离暂存和 Python 3.10 兼容,并同步依赖路径及中英文说明。
将 feat/h3-cross-hardware 与当前 main 合并;保留 H3 changelog 条目并追加在 main 新增条目之后,维持 append-only 契约。

Co-authored-by: Wenyao Gao <edwingao28@users.noreply.github.com>
将 H3 工作流、暂存功耗打包与 AMD 辅助脚本对齐到 inferencex-e2e/ 布局;保留 staged 包内的 infx/ 导入契约与手动派发路径。

Co-authored-by: Wenyao Gao <edwingao28@users.noreply.github.com>
补充 #2916 与 App #1193 的契约矩阵、CPU 验证结果与下一步决策说明。

Co-authored-by: Wenyao Gao <edwingao28@users.noreply.github.com>
新增 Wan 视频服务与部署证据投影,保留 H3 行为。
@github-actions

github-actions Bot commented Oct 5, 2026

Copy link
Copy Markdown
Contributor

Thanks for the contribution!

  • Review: If this PR changes files owned by someone other than a repository admin or @SemiAnalysisAI/core, ask one eligible CODEOWNER to complete the latest PR_REVIEW_CHECKLIST.md before contacting a core maintainer on Slack. Follow the template exactly, including As a PR reviewer and CODEOWNER, I have reviewed this and have, so sign-off verification triggers.
  • PR verification: Sweeps only run on labeled PRs. Add full-sweep-fail-fast (strongly recommended); use full-sweep-enabled only when matrix jobs should continue after a failure.
  • After merging: PR authors must ensure all GitHub Actions jobs pass. Transient failures often pass on rerun; see how to rerun failed jobs.
中文

感谢你的贡献!

  • **审阅:**如果 PR 修改的文件归属于仓库管理员及 @SemiAnalysisAI/core 之外的 CODEOWNER,请先联系一位有资格的 CODEOWNER 填写最新的 PR_REVIEW_CHECKLIST.md,再通过 Slack 联系核心维护者。必须严格遵循模板,并保留 As a PR reviewer and CODEOWNER, I have reviewed this and have,才能触发签核验证。
  • **PR 验证:**扫描仅在带有标签的 PR 上运行。强烈建议添加 full-sweep-fail-fast;仅当需要矩阵任务在失败后继续运行时才使用 full-sweep-enabled。
  • **合并后:**PR 作者必须确保所有 GitHub Actions 任务通过。临时性失败通常可以通过重新运行恢复;参见重新运行失败任务的说明。

@edwingao28
edwingao28 force-pushed the feat/h3-cross-hardware branch 6 times, most recently from ddc4127 to f7c7f58 Compare October 17, 2026 04:31
…aluator

Carries the MiniMax-H3 serving harness from #2916 onto main. The evaluator
gains srt-serve/srt-client/srt-finalize so a generic srt-slurm service runs
the diffusion server, each concurrency publishes its own measurement window,
and the dashboard bundle is assembled from the native power package.

中文:将 #2916 的 MiniMax-H3 serving harness 迁移到 main。evaluator 新增
srt-serve/srt-client/srt-finalize:由 srt-slurm generic service 启动 diffusion
server,每个并发度发布自己的测量窗口,并基于原生功耗包组装仪表板产物。
Adds the video-generation scenario to the vendor master configs and routes it
to a new srt-video launch path: the recipe binds a custom benchmark plus the
h3-server service, the job mounts the H3 package and HF cache, and
infx.results.video validates every concurrency against the native power
package with per-clip energy before the bundle is built.

中文:在厂商 master 配置中加入 video-generation scenario,并路由到新的
srt-video 启动路径:recipe 绑定 custom benchmark 与 h3-server service,作业挂载
H3 package 与 HF cache,infx.results.video 按并发度用原生功耗包校验并以每个
clip 归一化能耗,然后再组装产物。
@edwingao28
edwingao28 force-pushed the feat/h3-cross-hardware branch from f7c7f58 to 068d64f Compare October 17, 2026 04:42
e2e-tests.yml plans video-generation rows into a test-sweep-video job, the
benchmark template passes the H3 spec and concurrency list to the launcher and
uploads the h3-results bundle, and the standalone H3 workflows use the
harness's check_env.sh. One perf-changelog entry covers the four H3 keys.

中文:e2e-tests.yml 将 video-generation 行规划到 test-sweep-video 作业,benchmark
模板向 launcher 传递 H3 spec 与并发度列表并上传 h3-results 产物,独立的 H3
workflow 改用 harness 的 check_env.sh。perf-changelog 新增一条覆盖四个 H3 键。
@edwingao28
edwingao28 force-pushed the feat/h3-cross-hardware branch from 068d64f to 0cb4c84 Compare October 17, 2026 04:45
中文:将 PR #2916 的 H3 serving spec 原样复制为测试 fixture。
中文:新增 SGLang 与 vLLM-Omni 共用的 MiniMax-H3 workload spec 及 loader;生成的 plan 及其 plan_sha256 与 #2916 runner 一致。
中文:新增 CI,对 H3 workload spec 运行测试、lint 和格式检查。
中文:记录在 2x H100 上服务 MiniMax-H3 所用的 vLLM-Omni 镜像 pin、server 命令、并行配置与 attention backend,供 Phase B 使用。
@edwingao28
edwingao28 changed the base branch from feat/h3-cross-hardware to feat/videogenx-h3-vllm-omni October 18, 2026 09:23
@edwingao28 edwingao28 changed the title [VideoGenX] add Wan serving and deployment evidence / [VideoGenX] 新增 Wan serving 与部署证据 [VideoGenX] run matched Wan workloads on SGLang and vLLM-Omni Oct 18, 2026
Base automatically changed from feat/videogenx-h3-vllm-omni to feat/h3-cross-hardware October 29, 2026 12:19
@edwingao28
edwingao28 force-pushed the feat/h3-cross-hardware branch from 2c603bb to 5e2556d Compare October 29, 2026 20:11
@edwingao28 edwingao28 closed this Oct 29, 2026
@edwingao28 edwingao28 reopened this Oct 29, 2026
@edwingao28
edwingao28 force-pushed the feat/h3-cross-hardware branch 11 times, most recently from 6abe3b6 to a6db760 Compare November 3, 2026 09:53
Base automatically changed from feat/h3-cross-hardware to main November 5, 2026 21:45
@edwingao28 edwingao28 closed this Nov 12, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

1 participant