[None][test] Waive 6 failed cases for main in QA CI

[xinhe-nv]

Auto-generated Waive PR

Created by: TensorRT LLM CI Report (requested by qa@nvidia.com)
Target branch: main
Jenkins build: https://prod.blsm.nvidia.com/swqa-tensorrt-qa-test/job/LLM_FUNCTION_TEST/2117/
Bug(s): 6087946, 6112497

Waive entries added

accuracy/test_llm_api_pytorch.py::TestGemma3_1BInstruct::test_fp8_prequantized[torch_compile=False] SKIP (https://nvbugs/6087946)
accuracy/test_llm_api_pytorch.py::TestGemma3_1BInstruct::test_fp8_prequantized[torch_compile=True] SKIP (https://nvbugs/6087946)
accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_bfloat16_4gpus[pp4-attn_backend=TRTLLM-torch_compile=False] SKIP (https://nvbugs/6112497)

---

This PR was auto-generated by TensorRT LLM CI Report. Please review the waive entries before merging.

Summary by CodeRabbit

• Tests
  • Updated test skip list to exclude specific GPU acceleration and optimization test cases from validation runs.

[coderabbitai]

📝 Walkthrough

Walkthrough

The pull request adds three new test skip entries to the test waiver list, marking specific test cases for test_fp8_prequantized and test_bfloat16_4gpus as SKIP with associated NVIDIA bug links.

Changes

Cohort / File(s)	Summary
Test Waivers tests/integration/test_lists/waives.txt	Added three new SKIP entries for TestGemma3_1BInstruct::test_fp8_prequantized (with torch_compile variations) and TestLlama3_1_8BInstruct::test_bfloat16_4gpus (with pp4 and TRTLLM attention backend config), each with corresponding NV bug references.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Suggested reviewers

• jieli-matrix
• StanleySun639

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and specifically describes the main change: waiving 6 failed test cases for the main branch in QA CI.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
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.
Description check	✅ Passed	The pull request description clearly identifies the auto-generated nature, references the Jenkins build and associated bug links, and lists all waive entries being added to the test file.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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

[crazydemo]

/bot run --skip-test

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@tests/integration/test_lists/waives.txt`:
- Around line 433-435: The PR added only 3 waive entries but the objective
requires 6; add the three missing waive lines to the same file so all six failed
cases are waived: add entries for
accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-FP8-True]
SKIP (https://nvbugs/6112502),
accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-trtllm]
SKIP (https://nvbugs/6112503), and
accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] SKIP
(https://nvbugs/5921674) using the same formatting as the existing waive entries
so the PR meets the "Waive 6 failed cases" objective.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 29af17b0-2b2d-48a5-a6a1-da017f1f43af

📥 Commits

Reviewing files that changed from the base of the PR and between 04b454e and 1c3bdca.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

[coderabbitai]

⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Description: Search for the missing waive entries in test lists

echo "=== Searching for missing waive entries ==="

# Search for the missing test cases
rg -n "nvidia_Llama-3.1-8B-Instruct-FP8-True" tests/integration/test_lists/
rg -n "TestNemotronNanoV3.*fp8-4-trtllm" tests/integration/test_lists/
rg -n "TestNemotronV2.*test_fp8.*True" tests/integration/test_lists/

echo ""
echo "=== Checking for bug references ==="
rg -n "6112502|6112503" tests/integration/test_lists/

Repository: NVIDIA/TensorRT-LLM

Length of output: 1340

---

Add the missing 3 waive entries to complete the PR objective of waiving 6 failed cases.

The code changes add only 3 waive entries (lines 433–435), but the PR objectives state "Waive 6 failed cases" and list 6 specific cases. The following 3 entries are missing and should be added to waives.txt:

• accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-FP8-True] SKIP (https://nvbugs/6112502)
• accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-trtllm] SKIP (https://nvbugs/6112503)
• accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] SKIP (https://nvbugs/5921674)

The 3 entries present are correctly formatted per coding guidelines.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@tests/integration/test_lists/waives.txt` around lines 433 - 435, The PR added
only 3 waive entries but the objective requires 6; add the three missing waive
lines to the same file so all six failed cases are waived: add entries for
accuracy/test_llm_api_autodeploy.py::TestModelRegistryAccuracy::test_autodeploy_from_registry[nvidia_Llama-3.1-8B-Instruct-FP8-True]
SKIP (https://nvbugs/6112502),
accuracy/test_llm_api_autodeploy.py::TestNemotronNanoV3::test_accuracy[fp8-4-trtllm]
SKIP (https://nvbugs/6112503), and
accuracy/test_llm_api_autodeploy.py::TestNemotronV2::test_fp8[True] SKIP
(https://nvbugs/5921674) using the same formatting as the existing waive entries
so the PR meets the "Waive 6 failed cases" objective.

[crazydemo]

merge to #13504