[None][test] Update LLM performance test cases to use increased input/output lengths

[yufeiwu-nv]

Modified the llm_perf_core.yml file to replace several test cases with updated input/output lengths of 8000, enhancing the performance testing coverage for various models. Removed outdated test cases and ensured consistency across the benchmarks.

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

Summary by CodeRabbit

• Tests
  • Updated performance test coverage to include longer input and output lengths across multiple GPU and compute configurations.
  • Adjusted throughput and latency scenarios to reflect the updated benchmark parameters.
  • Added streaming performance coverage for the Nemotron Nano 12B v2 model.
  • Refined batch-size and memory-utilization scenarios for the expanded workloads.

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: 85b8f7ea-196d-4208-ae7d-8f47e9da83c1

📥 Commits

Reviewing files that changed from the base of the PR and between 29f4552 and 19a2122.

📒 Files selected for processing (1)

• tests/integration/test_lists/qa/llm_perf_core.yml

---

📝 Walkthrough

Walkthrough

The QA performance matrix replaces multiple shorter benchmark parameterizations with 8000,1000 input/output lengths across GPU condition groups, adjusts throughput and latency settings, and adds a BF16 streaming case for nemotron_nano_12b_v2.

Changes

LLM Performance Matrix

Layer / File(s)	Summary
Baseline GPU group updates tests/integration/test_lists/qa/llm_perf_core.yml	Sections 1–3 update Qwen and Nemotron BF16 benchmark entries to longer input/output lengths while retaining or adjusting latency and throughput parameters.
Shared long-context benchmark matrix tests/integration/test_lists/qa/llm_perf_core.yml	Sections 4–6 update BF16 and FP8 model cases, KV and scheduling parameters, and add a BF16 streaming test for nemotron_nano_12b_v2.
High-end GPU benchmark updates tests/integration/test_lists/qa/llm_perf_core.yml	Sections 7–10 update DeepSeek, Kimi, GLM, MiniMax, Qwen, Nemotron, and Llama benchmark variants to use longer input/output lengths.

Estimated code review effort: 4 (Complex) | ~45 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#16535: Updates integration performance-test registrations by replacing short 1k1k parameterizations with longer-length variants.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description is mostly template text and leaves the required Description and Test Coverage sections empty.	Replace the template text with a short issue summary, the change made, and concrete test coverage; keep the PR checklist completed.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and accurately reflects the performance test matrix update.
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.}

[yufeiwu-nv]

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

[tensorrt-cicd]

PR_Github #59941 [ skip ] triggered by Bot. Commit: 19a2122 Link to invocation

[tensorrt-cicd]

PR_Github #59941 [ skip ] completed with state SUCCESS. Commit: 19a2122
Skipping testing for commit 19a2122

Link to invocation