perf(dsv41flash): tune B200 vLLM AgentX DEP2/DEP4 and add TP2 c1 and DEP2 c128 host-KV rows / 调优 B200 vLLM AgentX DEP2/DEP4 并新增 TP2 c1 与 DEP2 c128 主机 KV 行

[hshrivastava-droid]

Summary

Tunes the B200 vLLM AgentX recipe for dsv41flash-fp4-b200-vllm-agentic-dspark and adds two new rows.

• Config: dsv41flash-fp4-b200-vllm-agentic-dspark
• Image: vllm/vllm-openai:nightly-dev-x86_64-cu130-ac9126e58aa7 (unchanged)

AI model disclosure

• Model/version: Claude Sonnet 5.5 (claude-sonnet-5-5)
• Role: prepared the PR

PR checklist and change type

Type of Change

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

Checklist

• I have completed the AI model disclosure and kept it current
• 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
• If this PR can affect benchmark performance or adds or modifies a recipe, it carries exactly one primary sweep label
• Before merging via reuse, an authorized maintainer has commented /use <run_id> on this PR

中文

调优 dsv41flash-fp4-b200-vllm-agentic-dspark 的 B200 vLLM AgentX recipe，并新增两行。配置：dsv41flash-fp4-b200-vllm-agentic-dspark；镜像：vllm/vllm-openai:nightly-dev-x86_64-cu130-ac9126e58aa7（不变）。改动类型：配置变更。

[github-actions]

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]

🔴 This new changelog entry will make the perf-changelog CI check fail for this PR because it was appended without the required blank-line separator from the prior entry. Line 9412 (pr-link: .../pull/3603) is immediately followed by - config-keys: at 9413 with no empty line between them, unlike every other entry boundary in this file. Fix: insert one blank line between the end of the pull/3603 entry and the new - config-keys: block, matching the separator pattern used throughout the rest of perf-changelog.yaml.

Why this was flagged

infx/workflows/validate_perf_changelog.py:validate_raw_change (lines 198-236) checks appended changelog bytes: since base_raw (the file before this PR) ends with a single \n (not \n\n), it requires the appended suffix to start with \n- config-keys: (one empty separator line), per line 213. The diff instead appends - config-keys: directly after pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/3603\n with zero blank lines, so suffix.startswith(expected_start) is false and the function raises ChangelogValidationError: 'new changelog entries must be separated from history by one empty line and appended at the end'. On the base branch every prior append has this blank line (verified: every - config-keys: line in the file is preceded by a blank line except this new one), so CI's changelog validation step, which gates merges per AGENTS.md's append-only/byte-sensitive rule, fails for this PR until the separator is restored.

Verification: The new changelog entry is appended without the required blank-line separator. In perf-changelog.yaml line 9412 ( pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/3603) is immediately followed by - config-keys: at line 9413 with no empty line between them, unlike every other entry boundary in the file.

[github-actions]

View unofficial run (performance): https://inferencex.semianalysis.com/inference?unofficialRun=38008138803

View unofficial run (accuracy): https://inferencex.semianalysis.com/evaluation?unofficialRun=38008138803