[None][test] amend for qa weekly core test list

[ruodil]

Summary by CodeRabbit

• Tests
  • Added performance testing for Llama v3.3 8B model with multiple precision formats (bf16, fp8) and streaming capabilities.
  • Updated performance test suite to focus on Llama v3.3 8B, Nemotron Nano 12B v2, and updated DeepSeek R1 configurations across various GPU setups and inference scenarios.

Description

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.

• 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.

[coderabbitai]

📝 Walkthrough

Walkthrough

A new model key mapping for llama_v3.3_8b was added to the test performance configuration. The test matrix was updated to replace numerous model/precision test invocations with new variants focused on llama_v3.3_8b, nemotron_nano_12b_v2, gpt_oss_20b_fp4, and deepseek_r1_0528_fp4.

Changes

Cohort / File(s)	Summary
Model Configuration Addition tests/integration/defs/perf/test_perf.py	Added new model path mapping: llama_v3.3_8b → llama-models-v3/llama-v3-8b-instruct-hf.
Test Matrix Restructuring tests/integration/test_lists/qa/llm_perf_core.yml	Removed coverage for multiple models (bielik_11b_v2.2, mistral_small_v3.1_24b, phi_4_multimodal_instruct, llama_v3.1 variants, starcoder2, mixtral_8x7b_v0.1, deepseek_v3_lite, and others); added/expanded test invocations for llama_v3.3_8b (bf16, streaming, fp8), nemotron_nano_12b_v2 (bf16, streaming, throughput), gpt_oss_20b_fp4, and deepseek_r1_0528_fp4 (including chunked prefill and large-GPU variants).

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is entirely composed of template boilerplate with no actual implementation details, rationale, or test coverage information provided by the author.	Fill in the Description section with what changes were made and why, the Test Coverage section with relevant test information, and ensure the PR Checklist is properly addressed.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly identifies the changes as amendments to the QA weekly core test list, which aligns with the changeset that updates test configurations and adds a new model mapping.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ 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.}

[coderabbitai]

Actionable comments posted: 2

🧹 Nitpick comments (1)

tests/integration/defs/perf/test_perf.py (1)

1-1: Update copyright year to 2026.

The file is being modified in 2026, so the copyright year should be updated from 2022-2025 to 2022-2026. As per coding guidelines: "update year on modified files".

Proposed fix

-# SPDX-FileCopyrightText: Copyright (c) 2022-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-FileCopyrightText: Copyright (c) 2022-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.

🤖 Prompt for AI Agents

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

In `@tests/integration/defs/perf/test_perf.py` at line 1, Update the SPDX
copyright header string "SPDX-FileCopyrightText: Copyright (c) 2022-2025 NVIDIA
CORPORATION & AFFILIATES. All rights reserved." to reflect 2026 (change
2022-2025 to 2022-2026) so the file's copyright year matches the modification
year; locate and replace that exact SPDX header line in the file.

🤖 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/defs/perf/test_perf.py`:
- Line 56: The entry for the model key "llama_v3.3_8b" uses the wrong path
"llama-models-v3/llama-v3-8b-instruct-hf"; update this value to follow the v3.3
naming convention by replacing it with "llama-3.3-models/Llama-3.3-8B-Instruct"
so it matches the pattern used by other v3.3 entries like
"llama_v3.3_70b_instruct".

In `@tests/integration/test_lists/qa/llm_perf_core.yml`:
- Around line 29-31: The YAML references model variants
(llama_v3.3_8b_instruct_fp8, llama_v3.3_8b_instruct, llama_v3.3_8b_instruct_fp4)
that are not defined and cause KeyError; add these variants to your model lookup
(e.g., extend MODEL_PATH_DICT or HF_MODEL_PATH) with entries that point to the
existing llama_v3.3_8b model path (or correct HF path), or add them to
allowed_configs.py if your validation expects explicit allowed names; ensure
keys exactly match the YAML names and reuse the same path/value as
MODEL_PATH_DICT["llama_v3.3_8b"] (or canonical HF id) so the tests resolve
without changing other code paths.

---

Nitpick comments:
In `@tests/integration/defs/perf/test_perf.py`:
- Line 1: Update the SPDX copyright header string "SPDX-FileCopyrightText:
Copyright (c) 2022-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved."
to reflect 2026 (change 2022-2025 to 2022-2026) so the file's copyright year
matches the modification year; locate and replace that exact SPDX header line in
the file.

🪄 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: Pro Plus

Run ID: d352a92d-92c4-487c-906b-9d1bd95f6718

📥 Commits

Reviewing files that changed from the base of the PR and between 2a0bcb1 and 5e30994.

📒 Files selected for processing (2)

• tests/integration/defs/perf/test_perf.py
• tests/integration/test_lists/qa/llm_perf_core.yml

[ruodil]

/bot skip --comment "skip test as just modifying cases"

[tensorrt-cicd]

PR_Github #44359 [ skip ] triggered by Bot. Commit: edd89d2 Link to invocation

[tensorrt-cicd]

PR_Github #44359 [ skip ] completed with state SUCCESS. Commit: edd89d2
Skipping testing for commit edd89d2

Link to invocation

[ruodil]

/bot reuse-pipeline

[tensorrt-cicd]

PR_Github #44667 [ reuse-pipeline ] triggered by Bot. Commit: c93075c Link to invocation

[tensorrt-cicd]

PR_Github #44667 [ reuse-pipeline ] completed with state SUCCESS. Commit: c93075c
Can't reuse PR_Github #0 with status: UNKNOWN

Link to invocation

[yufeiwu-nv]

/bot skip --comment "only test list modify"

[tensorrt-cicd]

PR_Github #44691 [ skip ] triggered by Bot. Commit: 9acacc9 Link to invocation

[tensorrt-cicd]

PR_Github #44691 [ skip ] completed with state SUCCESS. Commit: 9acacc9
Skipping testing for commit 9acacc9

Link to invocation