[TRTLLM-15102][fix] Prune VILA family models

[xinhe-nv]

…nctional and unit tests

Remove out-of-scope VILA, VILA1.5-3b, and NVILA-8B coverage per Test Scope Review 2026.7, matching the LLaVA-Next prune pattern.

Dev Engineer Review

• Removed out-of-scope VILA, VILA1.5-3B, and NVILA-8B accuracy and unit tests.
• Removed matching MMMU accuracy references and the VILA-specific model-root override.
• Removed the VILA1.5-3B attention configuration from the standard sweep.
• Updated QA, CI, and waiver lists to remove the affected tests.
• Changes are consistent with the stated test-scope pruning objective.
• Test-list formats and removed entries are consistent. No duplicate or unintended additions are present.

QA Engineer Review

• Removed integration test classes:
  • TestNVILA_8B
  • TestVILA1_5_3B
• Removed unit test module tests/unittest/_torch/modeling/test_modeling_vila.py, including VILA model tests and multimodal token-alignment tests.
• Removed related test-list entries from:
  • tests/integration/test_lists/qa/llm_function_core.txt
  • tests/integration/test_lists/qa/llm_function_rtx6k.txt
  • tests/integration/test_lists/test-db/l0_l40s.yml
  • tests/integration/test_lists/waives.txt
• The removed integration tests are also removed from their CI and manual QA lists.
• The removed unit tests had no corresponding entries in the listed CI or manual QA files.
• Verdict: sufficient.

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)

• 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]

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: 5a9974c0-f572-408e-8274-37b19a3739ea

📥 Commits

Reviewing files that changed from the base of the PR and between 0ab8c0c and 6ce348e.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

🚧 Files skipped from review as they are similar to previous changes (1)

• tests/integration/test_lists/waives.txt

---

Walkthrough

The PR removes VILA and NVILA multimodal accuracy tests, unit-test configuration, accuracy references, a fixture override, and a pre-merge test entry.

Changes

VILA and NVILA coverage cleanup

Layer / File(s)	Summary
Remove model-specific tests and references tests/integration/defs/accuracy/..., tests/unittest/_torch/attention/model_attn_config.py	The MMMU accuracy entries and multimodal accuracy classes were removed. The VILA attention configuration was also removed.
Clean test-suite wiring tests/integration/defs/conftest.py, tests/integration/test_lists/test-db/l0_l40s.yml	The VILA model-root override and pre-merge modeling test entry were removed.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to 6ce34

This change removes out-of-scope VILA-family test coverage and related test references without indicating a product or runtime behavior change. No actionable merge-blocking risk remains beyond normal checks and review.

Suggested reviewers: niukuo, crazydemo, perkzzheng, nv-guomingz, brnguyen2

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description states the pruning scope and reason but leaves Test Coverage and most checklist items incomplete.	Add the relevant tests and CI runs under Test Coverage, then complete the applicable checklist items.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the removal of VILA family model coverage from the test suite.
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.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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

[xinhe-nv]

/bot run --stage-list "l40s"

[tensorrt-cicd]

PR_Github #66168 [ run ] triggered by Bot. Commit: fb0a161 Link to invocation

[tensorrt-cicd]

PR_Github #66168 [ run ] completed with state FAILURE. Commit: fb0a161
/LLM/main/L0_MergeRequest_PR pipeline #53842 (Partly Tested) 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

[xinhe-nv]

/bot run --stage-list "L40S-*"

[xinhe-nv]

/bot run --stage-list "L40S-*"

[tensorrt-cicd]

PR_Github #66224 [ run ] triggered by Bot. Commit: 0ab8c0c Link to invocation

[pengbowang-nv]

Attention part change LGTM

[tensorrt-cicd]

PR_Github #66224 [ run ] completed with state SUCCESS. Commit: 0ab8c0c
/LLM/main/L0_MergeRequest_PR pipeline #53895 (Partly Tested) completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

CI Report

Link to invocation

[xinhe-nv]

/bot reuse-pipeline

[tensorrt-cicd]

PR_Github #66285 [ reuse-pipeline ] triggered by Bot. Commit: 6ce348e Link to invocation

[tensorrt-cicd]

PR_Github #66285 [ reuse-pipeline ] completed with state SUCCESS. Commit: 6ce348e
Reusing PR_Github #66224 (Partly Tested) for commit 6ce348e

Link to invocation