Skip to content

[Klaud Cold] minimaxm31flash-fp4-b300-sglang-agentic-dspark: day-zero MiniMax-M3.1-Flash B300 SGLang recipe - #4441

Merged
Oseltamivir merged 4 commits into
mainfrom
feat/minimaxm31flash-b300-sglang
Oct 25, 2026
Merged

Oseltamivir merged 4 commits into
mainfrom
feat/minimaxm31flash-b300-sglang

Conversation

@Oseltamivir

Copy link
Copy Markdown
Collaborator

Summary

Day-zero MiniMax-M3.1-Flash (MiniMaxAI/MiniMax-M3.1-Flash, NVFP4) AgentX on B300 with SGLang, following the SGLang cookbook: https://docs.sglang.io/cookbook/autoregressive/MiniMax/MiniMax-M3.1 (config source: docs/src/snippets/configs/MiniMaxAI/minimax-m3.1.jsx).

  • Image: lmsysorg/sglang:dev-minimax-m31-flash, pinned by digest (verified on Docker Hub 2026-10-24).
  • Serve flags: the cookbook's B300 Balanced (TP4, TRT-LLM routed MoE) for conc 1–32 and High-Throughput (TP8, attention DP2 via --attn-dp-size 2, EP8 MegaMoE) for conc 64–256. The matrix row is tp: 8, ep: 8 with dp-attn false, because the matrix's dp-attn means full DP attention.
  • Common flags: --speculative-algorithm DSPARK --speculative-use-block-verification, --cuda-graph-backend-prefill breakable, minimax-m3 reasoning and tool parsers, --tokenizer-worker-num 8.
  • Speculative decoding: seven-token DSpark (--speculative-dspark-block-size 7, the bundled draft's dspark_block_size). It's set explicitly so throughput runs resolve synthetic acceptance from minimaxm31flash_dspark.yaml ([Klaud Cold] Add MiniMax-M3.1-Flash DSpark golden AL curve and SPEED-Bench collector / [Klaud Cold] 新增 MiniMax-M3.1-Flash DSpark 黄金 AL 曲线与 SPEED-Bench 采集脚本 #4425): AL 4.04 at 7 tokens, thinking on (adaptive, the default).
  • Scheduler batch: the TP4 arm uses 2× CONC for AgentX subagent fan-out, capped at the cookbook's 128, with the decode graph batch matching it.
  • Checked locally: infx generate binds every matrix point to exactly one recipe variant, with SGLANG_SIMULATE_ACC_LEN=4.04.

vLLM siblings: #4419 (GB300), #4420 (MI355X), #4421 (B300), #4422 (H100), #4423 (H200), #4426/#4431 (B200), #4433 (GB200).

…-Flash SGLang AgentX on B300

Serve flags follow the SGLang cookbook's MiniMax-M3.1 cells with the day-zero
dev-minimax-m31-flash image. Seven-token DSpark with synthetic acceptance from
the minimaxm31flash_dspark golden curve.
@Oseltamivir
Oseltamivir requested a review from a team October 23, 2026 17:37
@Oseltamivir Oseltamivir added the full-sweep-fail-fast Full sweep with canary gate; first failure cancels the rest of that matrix (recommended) label Oct 23, 2026
@github-actions

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 run only when the PR appends an inferencex-e2e/perf-changelog.yaml entry and carries exactly one primary label: full-sweep-fail-fast (strongly recommended; canary plus per-matrix fail-fast), full-sweep-enabled (canary; matrix jobs continue after a failure), or non-canary-full-sweep-enabled (no canary or fail-fast). The modifiers all-evals, evals-only, and agentx-fast require a primary label. On fork PRs, a maintainer applies the label. See sweep labels and reuse.
  • 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:**只有当 PR 在 inferencex-e2e/perf-changelog.yaml 末尾追加了条目,并且恰好带有一个主标签时,才会运行扫描:full-sweep-fail-fast(强烈推荐;canary 加逐矩阵 fail-fast)、full-sweep-enabled(有 canary;矩阵任务在失败后继续运行)或 non-canary-full-sweep-enabled(无 canary,也无 fail-fast)。修饰标签 all-evals、evals-only 和 agentx-fast 必须与主标签一起使用。fork PR 的标签由维护者添加。参见扫描标签与复用。
  • **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。

@claude claude Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Findings marked 🟡 are optional suggestions and need no follow-up push.

Comment thread inferencex-e2e/perf-changelog.yaml Outdated
- "Serve flags follow the cookbook's B300 Balanced (TP4) and High-Throughput (TP8, attention DP2, EP8): DSpark with block verification, breakable prefill graphs, minimax-m3 reasoning and tool parsers, 8 tokenizer workers."
- "Seven-token DSpark (the bundled draft's dspark_block_size); throughput runs use synthetic acceptance from the minimaxm31flash_dspark golden curve (AL 4.04 at 7 draft tokens, #4425)."
- "Search space: Balanced TP4 at conc 1-32; TP8 + EP8 with attention DP2 at conc 64-256."
pr-link: TBD

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

🔴 The changelog check will reject this PR because its new entry's pr-link: TBD is not an accepted placeholder. Merge-time canonicalization will also skip it. validate_added_pr_link (infx/workflows/validate_perf_changelog.py:147) only accepts XXX, https://github.com/SemiAnalysisAI/InferenceX/pull/XXX, or the PR's own URL. Fix: change the line to pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/XXX or to the real PR URL. Every other appended entry in the file uses one of those forms.

Why this was flagged

The PR appends an entry whose pr-link is TBD (inferencex-e2e/perf-changelog.yaml:11018). On PR runs, validate_added_pr_link (inferencex-e2e/infx/workflows/validate_perf_changelog.py:139-150) accepts only PR_LINK_PLACEHOLDERS = {"XXX", "https://github.com/SemiAnalysisAI/InferenceX/pull/XXX"} or the expected PR URL. Anything else raises new PR entry must use ... or an XXX placeholder; found 'TBD'. On main-branch runs it must fully match CANONICAL_PR_LINK, so TBD raises new main-branch entry has invalid pr-link. The changelog validation fails, which blocks the sweep and merge. The merge-time placeholder rewrite in prepare_perf_changelog_merge.py / merge_with_reuse.py only targets the XXX placeholders, so it will not fix TBD either. No earlier entry in the file uses TBD; the other placeholders are .../pull/XXX (e.g. lines 10697, 10813).

Verification: The new entry sets pr-link: TBD (inferencex-e2e/perf-changelog.yaml:11018). merge_with_reuse.py:631 calls canonicalize_changelog(pr, repo). At validate_perf_changelog.py:147, validate_added_pr_link raises new PR entry must use ... or an XXX placeholder; found 'TBD'. canonicalize_changelog has no try/except, so the merge aborts.

Comment on lines +9362 to +9367
# MiniMax-M3.1-Flash AgentX on B300 with SGLang: day-zero image from the SGLang
# cookbook (https://docs.sglang.io/cookbook/autoregressive/MiniMax/MiniMax-M3.1),
# manifest digest verified 2026-10-24. Serve flags follow the cookbook's
# B300 Balanced (TP4) and High-Throughput (TP8, attention DP2, EP8).
# Seven-token DSpark (bundled draft); synthetic acceptance follows
# infx/golden_al_distribution/minimaxm31flash_dspark.yaml.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

🟡 nit (optional): The new config and recipe add header and provenance comments, which the mandatory comment rules in AGENTS.md:64-65 forbid. sweep:^# (MiniMax-M3.1-Flash AgentX|TP4 Balanced cell|High-Throughput cell)|# Serve flags follow the cookbook For example, nvidia-master.yaml:9362-9367 describes the config and says "manifest digest verified 2026-10-24". It is also dated a day after the PR was opened. Fix: delete these block, section and provenance comments. Keep only short per-line technical reasons, such as the dp-attn note. The cookbook URL and verification details already live in the perf-changelog entry.

Why this was flagged

AGENTS.md:62-66 says the comment rules cover every file, recipes and configs included. It bans comments above a recipe, override or section, and comments that describe the config. It also bans comments saying where something came from. The diff adds a 6-line block above minimaxm31flash-fp4-b300-sglang-agentic-dspark (nvidia-master.yaml:9362-9367) that describes the topology and records provenance ("day-zero image from the SGLang cookbook", "manifest digest verified 2026-10-24"). The recipe adds section comments above the override groups (agentic.yaml:51-52 and 129-131) and a cookbook-source comment (agentic.yaml:24-25). Reviewers and the CODEOWNER checklist have to flag these, and the comments are new rule violations that the base does not have. The verification date in the master config is after the PR's open date (2026-10-23), so that provenance claim is also unreliable.

Verification: nit: this happens on every merge of these files as written. AGENTS.md "Code comments (mandatory)" says "No header comments" and "No history". The diff adds a 6-line block right above minimaxm31flash-fp4-b300-sglang-agentic-dspark: in nvidia-master.yaml that describes the config and its provenance. The recipe adds the same kind of comments. Nothing breaks at runtime.

Keep main's entries and re-add the B300 SGLang changelog and master-config blocks.
@github-actions

github-actions Bot commented Oct 24, 2026 •

Copy link
Copy Markdown
Contributor

@Oseltamivir

Copy link
Copy Markdown
Collaborator Author

/use 40065872492

@github-actions

github-actions Bot commented Oct 25, 2026 •

Copy link
Copy Markdown
Contributor

@Oseltamivir staged run 40065872492.

This run remains available across future /stage-results requests. Staging the same run ID again updates its staged data. Staging workflow

@Oseltamivir
Oseltamivir merged commit 0c1e51e into main Oct 25, 2026
26 checks passed
@Oseltamivir
Oseltamivir deleted the feat/minimaxm31flash-b300-sglang branch October 25, 2026 12:15
@github-actions

Copy link
Copy Markdown
Contributor

Ingest report — ✅ Results ingested into the production database.

Ingest workflow · Merge Ingest run · Benchmark source run

Category new duplicate skipped unmapped failed
benchmark 8 0 8 0 0
eval 1 1 1 0 0
trace 8 0 0 0 0
power 8 0 0 0 0

benchmarkFiles 9, rawBenchmarkRows 16, coveredByPerJobFiles 8, purgedPoints 0, failedRunRows 0, runStatsNew 0, runStatsDuplicate 0, evalSamplesNew 1319, gpuMetricSamplesNew 274092, powerPublicationPoints 8, changelogEntriesWritten 1

Counts cover only this run. Rows: benchmark_results, eval_results, agentic rows linked to a trace, gpu_metrics series. This comment updates if the ingest is rerun.

中文

Ingest 报告 — ✅ 结果已写入 production 数据库。

上表仅统计本次运行写入的数据,各行单位依次为 benchmark_results 行、eval_results 行、已关联 trace 的 agentic 行和 gpu_metrics 序列。重新运行 ingest 时本评论会原地更新。

ispobock added a commit that referenced this pull request Oct 25, 2026
The day-zero recipe from #4441 is now on main, so this entry describes the change on top of it.

#4441 的 day-zero 配置已合入 main,因此这条记录改为描述在其基础上的更新。
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

full-sweep-fail-fast Full sweep with canary gate; first failure cancels the rest of that matrix (recommended)

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

1 participant