[https://nvbugs/6402009][test] unwaive Qwen3-235B NVFP4 TRTLLM MoE test

[jiahanc]

Dev Engineer Review

• Removed the NVBug 6402009 waiver for TestQwen3_235B_A22B::test_nvfp4[latency_moe_trtllm].
• The test remains listed in test-db/l0_gb200_multi_nodes.yml with a 90-minute timeout and isolation.
• The change has the expected one-line scope.
• Manual validation and pre-commit checks passed.

Verdict: sufficient

QA Engineer Review

• Modified file: tests/integration/test_lists/waives.txt.
• No test-db/ or qa/ files were modified.
• Removed the waiver for TestQwen3_235B_A22B::test_nvfp4[latency_moe_trtllm].
• The test is covered by tests/integration/test_lists/test-db/l0_gb200_multi_nodes.yml.
• The test also appears in tests/integration/test_lists/qa/llm_function_core.txt.
• CBTS coverage data is unavailable.

Verdict: needs follow-up

Description

Test Coverage

• uvx pre-commit run --files tests/integration/test_lists/waives.txt (passed)
• Manual 8x B300 validation on f41b625: 1 passed; MMLU 85.770 >= 83.683; GSM8K 86.277 >= 82.577
• CI target: l0_gb200_multi_nodes.yml, TIMEOUT (90), ISOLATION

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

[jiahanc]

/bot run

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: ada667f2-86c1-410c-8ab1-1c5481fa90ba

📥 Commits

Reviewing files that changed from the base of the PR and between c845c18 and f4e2bd6.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

💤 Files with no reviewable changes (1)

• tests/integration/test_lists/waives.txt

Included review availability: Your plan provides up to 12 included reviews per hour; 11 remain after this review.

---

Walkthrough

Removed the skip waiver for TestQwen3_235B_A22::test_nvfp4[latency_moe_trtllm], previously tracked by NV bug 6402009.

Changes

Cohort / File(s)	Summary
Test waiver cleanup tests/integration/test_lists/waives.txt	Removes the skip waiver for the Qwen3 NVFP4 latency test.

Estimated code review effort: 1 (Trivial) | ~2 minutes

Merge Risk: ⚪ Minimal · up to f4e2b

This change re-enables the existing Qwen3-235B NVFP4 TRTLLM MoE accuracy test without modifying production behavior; no actionable merge-blocking risk remains beyond normal checks and review.

Possibly related PRs

• NVIDIA/TensorRT-LLM#15195: Updates the same integration waiver file for a different test.
• NVIDIA/TensorRT-LLM#16997: Removes a different Qwen test waiver from the same file.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description includes relevant test coverage and checklist information, but the required Description section is empty. It does not explain the issue or the solution.	Add a concise Description section that states why NVBug 6402009 no longer requires a waiver and that removing the waiver re-enables the specified test in the two-node GB200 L0 stage.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title identifies NVBug 6402009, uses a valid type marker, and clearly states that the Qwen3-235B NVFP4 TRTLLM MoE test waiver is being removed.

Full details: Docstring Coverage

Explanation

No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0 files.

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[tensorrt-cicd]

PR_Github #69868 [ run ] triggered by Bot. Commit: f4e2bd6 Link to invocation

[xxi-nv]

/bot run --extra-stage "GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-1, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-2, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-3"

_{Context: the un-waived test TestQwen3_235B_A22B::test_nvfp4[latency_moe_trtllm] is registered only in the stage: post_merge block of l0_gb200_multi_nodes.yml, so the ordinary pre-merge pipeline never executes it. Adding the three GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-* shards on top of the normal pre-merge run so the un-waive is actually validated.}

[tensorrt-cicd]

PR_Github #69873 Bot args parsing error: usage: /bot [-h]
{run,kill,skip,submit,reviewers,reuse-pipeline,reuse-review} ...
/bot: error: unrecognized arguments: _{Context: the un-waived test TestQwen3_235B_A22B::test_nvfp4[latency_moe_trtllm] is registered only in the stage: post_merge block of l0_gb200_multi_nodes.yml, so the ordinary pre-merge pipeline never executes it. Adding the three GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-* shards on top of the normal pre-merge run so the un-waive is actually validated.}

Link to invocation

[xxi-nv]

/bot run --extra-stage "GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-1, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-2, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-3"

[tensorrt-cicd]

PR_Github #69876 [ run ] triggered by Bot. Commit: f4e2bd6 Link to invocation

[tensorrt-cicd]

PR_Github #69868 [ run ] completed with state ABORTED. Commit: f4e2bd6

Link to invocation

[tensorrt-cicd]

PR_Github #69876 [ run ] completed with state SUCCESS. Commit: f4e2bd6
/LLM/main/L0_MergeRequest_PR pipeline #57163 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

• Please check the failed tests and fix your PR
• If you cannot view the failures, ask the CI triggerer to share details
• Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

[xxi-nv]

/bot run --stage-list "GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-1, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-2, GB200-8_GPUs-2_Nodes-PyTorch-Post-Merge-3" --disable-fail-fast

[tensorrt-cicd]

PR_Github #70214 [ run ] triggered by Bot. Commit: f4e2bd6 Link to invocation

[tensorrt-cicd]

PR_Github #70214 [ run ] completed with state SUCCESS. Commit: f4e2bd6
/LLM/main/L0_MergeRequest_PR pipeline #57471 (Partly Tested) completed with status: 'SUCCESS'

CI Report

Link to invocation

[xxi-nv]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #70262 [ run ] triggered by Bot. Commit: f4e2bd6 Link to invocation

[tensorrt-cicd]

PR_Github #70262 [ run ] completed with state SUCCESS. Commit: f4e2bd6
/LLM/main/L0_MergeRequest_PR pipeline #57509 completed with status: 'SUCCESS'

CI Report

Link to invocation