TensorRT-LLM [https://nvbugs/5680905][fix] Relax the MMLU accuracy requirement for DS-v3.2

Summary by CodeRabbit

Tests
- Re-enabled previously skipped DeepSeekV32 test cases.
Chores
- Updated accuracy reference values for DeepSeek-V3.2-Exp model.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

Description

Change the DS-V3.2 MMLU reference accuracy from 87.81 to 87.2.
Unwaive two tests
- accuracy/test_disaggregated_serving.py::TestDeepSeekV32Exp::test_auto_dtype[False]
- accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_nvfp4_multi_gpus[baseline]

Test Coverage

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)
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.
[x] Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

Nov 25 '25 08:11 lfr-0531

/bot run

Nov 25 '25 08:11 lfr-0531

📝 Walkthrough

Walkthrough

Two accuracy reference values for DeepSeek-V3.2-Exp in the MMLU benchmark are updated from 87.81 to 87.2, and two DeepSeekV32-related test skip waivers are removed from the exclusion list.

Changes

Cohort / File(s)	Summary
MMLU accuracy reference updates `tests/integration/defs/accuracy/references/mmlu.yaml`	Updated two accuracy entries for DeepSeek-V3.2-Exp (quant_algo NVFP4, spec_dec_algo MTP) from 87.81 to 87.2
Test waiver removals `tests/integration/test_lists/waives.txt`	Removed two SKIP entries: `accuracy/test_disaggregated_serving.py::TestDeepSeekV32Exp::test_auto_dtype[False]` and `accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_nvfp4_multi_gpus[baseline]`

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

Verify the accuracy reference value changes are intentional and correctly reflect the new benchmark results
Confirm that removing the test skip waivers aligns with bug fixes or test stability improvements

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: relaxing MMLU accuracy requirements for DeepSeek-V3.2, which aligns with the changes in the raw summary.
Description check	✅ Passed	The PR description provides specific details about the changes (MMLU accuracy update and test unwaiving) but lacks explanation of why these changes are necessary or what issue they address.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

🧪 Generate unit tests (beta)

[ ] Create PR with unit tests
[ ] Post copyable unit tests in a comment

📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 4742c130dbb4196d2be57a285d356a1f20a225ce and 36ca08f6d910c4941b9e1f32e9cddbc41fbc5ac4.

📒 Files selected for processing (2)

tests/integration/defs/accuracy/references/mmlu.yaml (1 hunks)
tests/integration/test_lists/waives.txt (0 hunks)

💤 Files with no reviewable changes (1)

tests/integration/test_lists/waives.txt

🧰 Additional context used

🧠 Learnings (1)

📓 Common learnings

Learnt from: djns99
Repo: NVIDIA/TensorRT-LLM PR: 6915
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu:4010-4012
Timestamp: 2025-08-14T23:23:27.449Z
Learning: For MOE (Mixture of Experts) code reviews in TensorRT-LLM, avoid repeatedly suggesting finalize fusion validation checks and safety assertions. The user djns99 has indicated these suggestions are repetitive and unwanted across multiple MOE-related changes.

Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

GitHub Check: Pre-commit Check

🔇 Additional comments (2)

tests/integration/defs/accuracy/references/mmlu.yaml (2)

201-205: Missing file from review: waives.txt inconsistency with AI summary.

The AI-generated summary mentions that two DeepSeekV32-related test skip waivers are removed from tests/integration/test_lists/waives.txt, but this file was not included in the files for review. Verify whether waives.txt was indeed modified as part of this PR or if the summary is inaccurate.

201-205: Accuracy values updated consistently for DeepSeek-V3.2-Exp NVFP4 entries.

The two accuracy values (lines 202 and 205) for DeepSeek-V3.2-Exp with NVFP4 quantization are updated from 87.81 to 87.2, relaxing the accuracy requirement by 0.61 points. The changes are correctly formatted and consistently applied to both the base NVFP4 entry and the NVFP4+MTP speculative decoding variant.

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

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

Nov 25 '25 08:11 coderabbitai[bot]

PR_Github #25705 [ run ] triggered by Bot. Commit: 36ca08f

Nov 25 '25 08:11 tensorrt-cicd

PR_Github #25705 [ run ] completed with state FAILURE. Commit: 36ca08f /LLM/main/L0_MergeRequest_PR pipeline #19487 completed with status: 'FAILURE'

Nov 25 '25 08:11 tensorrt-cicd

/bot run

Nov 25 '25 09:11 lfr-0531

PR_Github #25720 [ run ] triggered by Bot. Commit: 36ca08f

Nov 25 '25 09:11 tensorrt-cicd

PR_Github #25720 [ run ] completed with state SUCCESS. Commit: 36ca08f /LLM/main/L0_MergeRequest_PR pipeline #19502 completed with status: 'SUCCESS' Pipeline passed with automatic retried tests. Check the rerun report for details.

Nov 25 '25 15:11 tensorrt-cicd