[Klaud Cold] Enact the August 3, 2026 scenario and precision deprecations / 执行 2026 年 8 月 3 日场景与精度下线

[functionstackx]

Summary

Today is Tuesday, August 4, 2026, so the August 3, 2026 block of the Deprecation Notice in MODELS.md is now past its last run day. This PR enacts the scenario and precision retirements from that block.

54 config keys move out of the active NVIDIA and AMD master configs into configs/deprecated/, and their 45 benchmark scripts move into the sibling deprecated/ directories.

Group	Keys	What
minimaxm3-8k1k	39	Single-turn 8k1k (agentic coding stays active)
kimik2.5-agentic	6	Agentic coding (8k1k stays active until August 6)
qwen3.5-bf16	9	All bf16 recipes, every scenario, both platforms

Archive files: configs/deprecated/{nvidia,amd}-{minimaxm3-8k1k,kimik2.5-agentic,qwen3.5-bf16}-master.yaml.

Not in this PR: the speculative-decoding A/B retirements

The second table in the August 3 block retires the non-spec-decode agentic arm of five models. That round is deliberately deferred, because the replacement arms do not exist yet:

• MiniMax-M3 and GLM-5.2 would be left with no active config at all — MiniMax-M3's EAGLE3 agentic arm is still in flight in [AMD] [WIP] [AGENTX] MiniMax-M3 Support on MI355X with MTP #2487, and no GLM-5.2 MTP arm has landed.
• DeepSeek-V4-Pro and Qwen3.5 would lose all AMD agentic coverage and (for dsv4) all SGLang agentic coverage; neither has an MTP sibling on those platform/engine combinations.
• Kimi-K3 keeps both arms.

Every non-spec-decode agentic config therefore stays active here. MODELS.md carries an explicit "Status: not yet enacted" note under that table with the reasoning, and the dsv4 / glm5.2 matrix rows no longer advertise MTP-only coverage that is not actually in force.

Verification

• Exact partition. For each platform, active ∪ archived == origin/main at the key level, active ∩ archived == ∅, and every entry's parsed value is byte-identical to origin/main. Blocks were moved as verbatim text, so comments and formatting are preserved. NVIDIA 157 → 125 active + 32 archived; AMD 78 → 56 active + 22 archived.
• Non-spec agentic arms confirmed still active — spot-checked dsv4-fp4-b200-vllm-agentic, glm5.2-fp4-b300-sglang-agentic, minimaxm3-fp4-mi355x-vllm-agentic, minimaxm3-fp8-h100-vllm-agentic, qwen3.5-fp8-mi355x-sglang-agentic, kimik3-fp4-b300-vllm-agentic.
• No active config lost its script. Cross-checked every moved script's (prefix, precision, runner-family, scenario) fingerprint against all 181 surviving configs — zero collisions.
• Moved scripts still resolve. Relative-path depth bumped in all 45 moved scripts (benchmark_lib.sh source, INFERENCEX_ROOT, chat_templates/). Every relative reference resolves to an existing file and all pass bash -n.
• No stale references. Repo-wide grep for the old script paths found 3 comment references (kimik2.5_fp4_b200.sh, models_vllm.yaml, KLAUD_DEBUG.md); all repointed. configs/ci-priority.yaml and configs/CONFIGS.md reference none of the deprecated keys.
• Following convention from Deprecate 1k1k benchmark configurations #2263 / [Klaud Cold] Deprecate GLM-5 / GLM-5.1 benchmark configs / 弃用 GLM-5 / GLM-5.1 基准测试配置 #2276 / [on june 22, 2026, current gptoss submiters were given notice that it will be deprepcated on july 6th, and were given an chance for comment period and everyone reacted notice] Deprecate gpt-oss-120b benchmark configs / 弃用 gpt-oss-120b 基准测试配置 #2101: launcher model-path routing and srt-slurm-recipes/ YAMLs are left in place for deprecated models, and no perf-changelog.yaml entry is added.

中文说明

今日为 2026 年 8 月 4 日（星期二），MODELS.md 弃用公告中 2026 年 8 月 3 日 一节已过最后运行日。本 PR 执行其中的场景与精度下线：54 个配置键从启用的 NVIDIA 与 AMD master 配置移入 configs/deprecated/，其 45 个基准测试脚本移入同级 deprecated/ 目录。

投机解码 A/B 下线本轮不执行：其替代分支尚未合入 —— 若此刻移除，MiniMax-M3 与 GLM-5.2 将没有任何启用配置，DeepSeek-V4-Pro 与 Qwen3.5 在 AMD 及 SGLang 上的智能体覆盖亦将全部消失。所有非投机解码智能体配置在本 PR 中保持启用，MODELS.md 已注明该表「尚未执行」及其原因。

[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]

Additional findings (outside current diff — PR may have been updated during review):

• 🔴 configs/deprecated/nvidia-agentic-non-spec-master.yaml:92-121 — configs/deprecated/nvidia-agentic-non-spec-master.yaml archives four GB300 dynamo-sglang agentic keys (qwen3.5-fp4-gb300-dynamo-sglang-agentic-agg/-disagg, dsv4-fp4-gb300-dynamo-sglang-agentic-agg/-disagg) and removes them from configs/nvidia-master.yaml, but every search-space row in all four uses spec-decoding: mtp — there is no non-MTP sibling for these keys.

  Extended reasoning...

  configs/deprecated/nvidia-agentic-non-spec-master.yaml is meant to hold only the non-speculative-decode arm of the dsv4/qwen3.5/glm5.2 agentic A/B pairs — its own header comment says so explicitly: "Deprecated non-speculative-decoding agentic-coding entries archived from nvidia-master.yaml ... the spec-decode arm is the better Pareto frontier, so only that arm is run and published." That contract is violated by four of the entries this PR moved into the file:

  • qwen3.5-fp4-gb300-dynamo-sglang-agentic-agg — both search-space rows set spec-decoding: "mtp" (lines ~105, 121 of the archive)
  • qwen3.5-fp4-gb300-dynamo-sglang-agentic-disagg — all 7 rows set spec-decoding: "mtp"
  • dsv4-fp4-gb300-dynamo-sglang-agentic-agg — its single row sets spec-decoding: mtp
  • dsv4-fp4-gb300-dynamo-sglang-agentic-disagg — all rows set spec-decoding: mtp

  These are the spec-decode (MTP) arm, not the non-spec arm the file is supposed to collect. Contrast with the genuinely non-spec entries correctly archived in the same file (e.g. dsv4-fp4-gb300-dynamo-vllm-agentic, the dsv4-fp4-gb200-dynamo-vllm-agentic-* entries), which all set spec-decoding: none/omit it.

  The Deprecation Notice in MODELS.md that this PR itself enacts is explicit: for both qwen3.5 and dsv4, only the non-MTP agentic arm is deprecated; "Agentic coding, MTP" is the arm that "remains"/is published. Grepping the active configs/nvidia-master.yaml after this PR confirms none of these four gb300-dynamo-sglang-agentic keys remain active, and there is no like-for-like GB300 dynamo-sglang replacement — the block that stayed active for qwen3.5 GB300-class agentic is qwen3.5-fp4-b300-sglang-agentic-mtp, which is a different recipe (single-node b300, not multi-node gb300-dynamo-sglang), and dsv4 keeps only gb300-dynamo-vllm agentic MTP, not gb300-dynamo-sglang. So this PR silently drops real, currently-published GB300 dynamo-sglang MTP agentic coverage for both models with no replacement.

  The likely root cause: the migration script/author probably classified entries into "non-spec" vs "spec" buckets by key-name suffix (looking for a trailing -mtp) rather than by the actual spec-decoding field inside the search-space rows. These four keys don't have an -mtp suffix in their names (unlike e.g. qwen3.5-fp4-b300-sglang-agentic-mtp), even though their contents are MTP recipes — so they fell through the suffix-based filter into the non-spec archive.

  Proof, step by step:

  1. Open configs/deprecated/nvidia-agentic-non-spec-master.yaml in this diff and find qwen3.5-fp4-gb300-dynamo-sglang-agentic-agg:.
  2. Its search-space rows read - spec-decoding: "mtp" ... conc-list: [1] and - spec-decoding: "mtp" ... conc-list: [96] — both explicitly MTP, not "none".
  3. Compare to the file's own header: "only the spec-decode arm is run and published" is deprecated-file language for removing the non-spec arm — but these rows ARE the spec-decode arm.
  4. Check configs/nvidia-master.yaml post-PR: grep gb300-dynamo-sglang-agentic returns nothing — the key is gone from the active config with no replacement of the same platform/engine/multinode shape.
  5. Check MODELS.md's own updated matrix in this PR: for qwen3.5, "Agentic coding (MTP only from 2026-08-03)" is listed as the active/remains scenario, and for dsv4 the same. Removing the MTP entries directly contradicts what the PR's own documentation says should stay active.

  Impact: this is a benchmark-coverage regression, not a crash — no user-facing failure occurs, but real, currently-published GB300 multi-node dynamo-sglang MTP agentic benchmark coverage for two models (qwen3.5 and dsv4) silently disappears from the sweep, with the archive's header and MODELS.md's matrix now factually describing coverage that no longer exists on that platform.

  Fix: move these four keys (and their search-space content) back into configs/nvidia-master.yaml as active configs — they are the arm the Deprecation Notice says to keep — or, if their removal was actually intended for a different reason, update MODELS.md and the archive header to correctly describe why the GB300 dynamo-sglang MTP arm no longer runs.

[claude]

🟡 The self-referential PR link uses a literal PRNUM placeholder instead of 2493 in both MODELS.md (lines 15, 56, 57, 61) and the mirrored lines in MODELS_zh.md, producing 8 broken /pull/PRNUM links (404) in the docs that are this PR's own deliverable. Fix with a simple find-and-replace of PRNUM → 2493 in both files.

Extended reasoning...

This PR's whole purpose is to record the enactment of the August 3, 2026 deprecation notice and link back to the PR that did it. Four places in MODELS.md (lines 15, 56, 57, 61) and the four mirrored lines in MODELS_zh.md write [#PRNUM](https://github.com/SemiAnalysisAI/InferenceX/pull/PRNUM) where the actual PR number, 2493, should be. This is a leftover template placeholder — the author correctly filled in the real numbers for every other PR reference on those lines (e.g. #2268, #1724, #734), but couldn't have known this PR's own number ahead of time and never went back to substitute it once the PR was opened.

I verified this against the actual committed file content, not the PR diff view: grep -n PRNUM MODELS.md MODELS_zh.md on the current working tree returns all 8 occurrences, and git show HEAD:MODELS.md confirms the same text is present in the commit that shipped this change (5b5606e, "docs(MODELS): record the August 3, 2026 deprecations as enacted"). The diff shown during review substitutes the placeholder with #2493 for display purposes, which is why it can look already-fixed there — but the file actually on disk/in HEAD still has the raw placeholder.

Concretely, anyone who clicks the "Enacted on 2026-08-04 in #PRNUM" link at MODELS.md:15, or either of the "no active config as of 2026-08-04 (#PRNUM)" callouts at lines 56-57, or the "Agentic coding (removed 2026-08-04, #PRNUM)" note at line 61, lands on https://github.com/SemiAnalysisAI/InferenceX/pull/PRNUM, which is not a valid PR number and 404s. Same for the four mirrored Chinese-language lines in MODELS_zh.md. Since the whole point of these notes is to let a future reader jump straight to the PR that performed the archival/removal, a dead link defeats that purpose, even though it doesn't break any code, CI, or benchmark behavior.

Fix is mechanical: replace PRNUM with 2493 in both files (8 substitutions total, matching the #2493 display already shown in the PR's own diff view). Since this is a documentation-only defect with no functional or CI impact, I'm marking it a nit rather than a blocking issue — but it should be fixed before merge since it's trivial and the whole point of the change is to leave working self-referential links.