[https://nvbugs/6221621][test] Update trust_remote to nemotron and phi4 models

[yufeiwu-nv]

Summary by CodeRabbit

• Tests
  • Consolidated model test configuration to reduce duplication and improve maintainability
  • Extended validation coverage for additional model variants

Description

Included additional models "nemotron_nano_12b_v2", "phi_4_multimodal_instruct", "phi_4_multimodal_instruct_fp4", and "phi_4_multimodal_instruct_fp8" to the TRUST_REMOTE_CODE_MODELS dictionary to enhance testing coverage.

Signed-off-by: yufeiwu-nv 230315618+yufeiwu-nv@users.noreply.github.com

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)

• 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.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR refactors model-path configuration for performance integration tests by extracting shared lookup tables (MODEL_PATH_DICT, HF_MODEL_PATH, LORA_MODEL_PATH) into a new _model_paths.py module, updating test_perf.py and test_perf_sanity.py to import from it, extending trust-remote-code model support, and adjusting test waives for B200 quantization tests.

Changes

Perf test model path refactoring

Layer / File(s)	Summary
Shared model path constants module tests/integration/defs/perf/_model_paths.py	New module exports MODEL_PATH_DICT (model keys to local/synced paths), HF_MODEL_PATH (model keys to HuggingFace identifiers), and LORA_MODEL_PATH (LoRA adapter mapping) for use across perf tests.
Refactor test_perf.py to import shared paths and extend trust-remote-code models tests/integration/defs/perf/test_perf.py	Copyright year updated to 2026. Model-path dictionaries moved from inline definitions to imports from ._model_paths. TRUST_REMOTE_CODE_MODELS extended to include nemotron_nano_12b_v2, phi_4_multimodal_instruct, phi_4_multimodal_instruct_fp4, and phi_4_multimodal_instruct_fp8.
Refactor test_perf_sanity.py with base model path override tests/integration/defs/perf/test_perf_sanity.py	Copyright year updated to 2022-2026. Imports shared MODEL_PATH_DICT base from ._model_paths, overrides llama_v3.3_70b_instruct_fp4 with historical path value, and adds process_and_upload_test_results import.
Update test waives for B200 perf tests tests/integration/test_lists/waives.txt	B200 perf test waive entries restructured to retain quantization-related cases (quant:int8_sq_per_tensor, quant:int8_sq_per_token_channel, quant:w4a8_awq) while removing non-quantization waives from the same block.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#14393: Refactors the same perf test model-path configuration and adjusts corresponding test waives.

Suggested reviewers

• StanleySun639
• mzweilz

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is incomplete. It lacks PR title format and does not fully document the scope of changes made.	Add a properly formatted PR title following [type] format and complete the Description section explaining all changes (model path refactoring, new file creation, imports). Fill in the Test Coverage section with relevant test details.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main change: adding models to TRUST_REMOTE_CODE_MODELS with the NVBugs ticket reference.
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.

_{✏️ 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]

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

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

17-147: 🛠️ Refactor suggestion | 🟠 Major | ⚡ Quick win

Add __all__ to declare the module’s public interface.

This module is now a shared constants API, but it doesn’t declare exported names via __all__.

Proposed patch

 """Shared model path constants for perf and perf-sanity tests."""
 
+__all__ = ["MODEL_PATH_DICT", "HF_MODEL_PATH", "LORA_MODEL_PATH"]
+
 # Model PATH of local dir synced from internal LLM models repo
 MODEL_PATH_DICT = {

As per coding guidelines: "Despite the prohibition on wildcard imports, keep __all__ updated to document the public interface".

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@tests/integration/defs/perf/_model_paths.py` around lines 17 - 147, Add a
module-level __all__ listing the public constants to document and control the
exported API; include the three symbols MODEL_PATH_DICT, HF_MODEL_PATH, and
LORA_MODEL_PATH (i.e. set __all__ = ["MODEL_PATH_DICT", "HF_MODEL_PATH",
"LORA_MODEL_PATH"]) so imports/export behavior is explicit for callers of this
shared constants module.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Outside diff comments:
In `@tests/integration/defs/perf/_model_paths.py`:
- Around line 17-147: Add a module-level __all__ listing the public constants to
document and control the exported API; include the three symbols
MODEL_PATH_DICT, HF_MODEL_PATH, and LORA_MODEL_PATH (i.e. set __all__ =
["MODEL_PATH_DICT", "HF_MODEL_PATH", "LORA_MODEL_PATH"]) so imports/export
behavior is explicit for callers of this shared constants module.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 048d27f8-5e7b-4a4d-b2bd-d3da0f79c321

📥 Commits

Reviewing files that changed from the base of the PR and between 1f8312d and 7436798.

📒 Files selected for processing (4)

• tests/integration/defs/perf/_model_paths.py
• tests/integration/defs/perf/test_perf.py
• tests/integration/defs/perf/test_perf_sanity.py
• tests/integration/test_lists/waives.txt

💤 Files with no reviewable changes (1)

• tests/integration/test_lists/waives.txt

[yufeiwu-nv]

/bot skip --comment "test modify"

[tensorrt-cicd]

PR_Github #50387 [ skip ] triggered by Bot. Commit: 8fe688b Link to invocation

[tensorrt-cicd]

PR_Github #50387 [ skip ] completed with state SUCCESS. Commit: 8fe688b
Skipping testing for commit 8fe688b

Link to invocation

[yufeiwu-nv]

/bot skip --comment "test modify"

[tensorrt-cicd]

PR_Github #50466 [ skip ] triggered by Bot. Commit: 1b7b574 Link to invocation

[tensorrt-cicd]

PR_Github #50466 [ skip ] completed with state SUCCESS. Commit: 1b7b574
Skipping testing for commit 1b7b574

Link to invocation