[None][test] Add K25 EPLB

[chenfeiz0326]

Summary by CodeRabbit

• Tests
  • Updated performance sanity test configurations for multi-GPU and multi-node deployments on GB200 and GB300 systems.
  • Added new test entries covering various hardware configurations and parameter combinations.
  • Enhanced test configurations with load balancer settings for improved resource distribution.

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]

📝 Walkthrough

Walkthrough

This PR updates performance sanity test configurations to enable kimi-k25-thinking-fp4 model testing with a load_balancer eplb416 parameter across GB200 and GB300 hardware. Test registry entries are modified or added, and corresponding model configuration files define the new load_balancer section with num_slots: 416 and layer_updates_per_iter: 1.

Changes

Performance Sanity Tests with eplb416

Layer / File(s)	Summary
GB200 eplb416 test updates and config tests/integration/test_lists/test-db/l0_gb200_multi_gpus_perf_sanity.yml, tests/integration/test_lists/test-db/l0_gb200_multi_nodes_perf_sanity_ctx1_node1_gpu4_gen1_node8_gpu32.yml, tests/scripts/perf-sanity/disaggregated/gb200_kimi-k25-thinking-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb416_mtp3_ccb-NIXL.yaml	Test entries for kimi-k25-thinking-fp4 con1024/dep32/mtp3 are updated from eplb0 to eplb416. Corresponding load_balancer configuration with num_slots: 416 is added to the GB200 model config.
GB300 test registry expansion and config tests/integration/test_lists/test-db/l0_gb300_multi_gpus_perf_sanity.yml, tests/integration/test_lists/test-db/l0_gb300_multi_nodes_perf_sanity_ctx1_node1_gpu4_gen1_node1_gpu4.yml, tests/integration/test_lists/test-db/l0_gb300_multi_nodes_perf_sanity_ctx1_node1_gpu4_gen1_node2_gpu8.yml, tests/integration/test_lists/test-db/l0_gb300_multi_nodes_perf_sanity_ctx1_node1_gpu4_gen1_node8_gpu32.yml, tests/scripts/perf-sanity/disaggregated/gb300_kimi-k25-thinking-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb416_mtp3_ccb-NIXL.yaml	Multiple new test entries for kimi-k25-thinking-fp4 with updated parameterization are added to GB300 test configurations across single GPU, dual GPU, and multi-node GPU scenarios. Load_balancer configuration with num_slots: 416 is added to the GB300 model config.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Suggested reviewers

• fredricz-20070104
• ruodil
• HuiGao-NV

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is empty beyond the template boilerplate. No actual description of the changes, rationale, test coverage, or implementation details are provided.	Fill in the Description and Test Coverage sections with clear explanations of what changes were made, why they were necessary, and which tests validate these changes.

✅ Passed checks (4 passed)

Check name	Status	Explanation
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.
Title check	✅ Passed	The title clearly identifies the main change: adding EPLB (Elastic Pipeline Load Balancing) configuration for the K25 model in test files.

_{✏️ 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.}

[chenfeiz0326]

/bot run --post-merge --stage-list "GB200-4_GPUs-PyTorch-PerfSanity-Post-Merge-,GB300-4_GPUs-PyTorch-PerfSanity-Post-Merge-,GB200-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-,GB300-36_GPUs-9_Nodes-PyTorch-Disagg-PerfSanity-CTX1-NODE1-GPU4-GEN1-NODE8-GPU32-Post-Merge-"

[tensorrt-cicd]

PR_Github #52293 [ run ] triggered by Bot. Commit: c070a9a Link to invocation

[tensorrt-cicd]

PR_Github #52293 [ run ] completed with state FAILURE. Commit: c070a9a
/LLM/main/L0_MergeRequest_PR pipeline #41602 (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