[None][fix] Fix moe backend setting

[shuyixiong]

Summary by CodeRabbit

• Fix
  resolve_moe_cls unconditionally downgraded an unquantized (BF16) MoE from TRTLLMGenFusedMoE to CutlassFusedMoE, so moe_config.backend="TRTLLM" was silently ignored for BF16 checkpoints. This PR removes that downgrade so the user-requested TRTLLM backend is actually used on its supported BF16 path.
• Root cause
  The downgrade rule (if moe_cls == TRTLLMGenFusedMoE and not has_quant: moe_cls = CutlassFusedMoE) was originally added as a fallback: some BF16 cases lacked FlashInfer support, so it fell back to Cutlass for those. It was gated by _supports_flashinfer_bf16_routing_method(routing_method), which kept TRTLLMGenFusedMoE whenever the FlashInfer BF16 path was actually available. That guard was dropped in [TRTLLM-13052][feat] Enable TRTLLM moe backend for nemotron-h BF16 ckpt #14944, turning a conditional fallback into an unconditional downgrade that discards a valid TRTLLM selection.

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

Removed a fallback conditional in resolve_moe_cls that previously downgraded the selected MoE class from TRTLLMGenFusedMoE to CutlassFusedMoE when the effective quantization config had no applicable quantization. Backend selection now relies solely on get_moe_cls's output.

Changes

MoE Backend Selection

Layer / File(s)	Summary
Remove quant-based downgrade fallback tensorrt_llm/_torch/modules/fused_moe/create_moe.py	Removed the has_quant check and conditional reassignment that forced CutlassFusedMoE when the effective quant config had no applicable quantization, leaving get_moe_cls's selection as final.

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

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is still the template with no actual title, issue summary, or test coverage details.	Add a proper PR title plus short Description and Test Coverage sections that explain the problem, solution, and relevant tests.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title matches the PR's MOE backend selection fix and is specific enough to identify the main change.
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.}

[shuyixiong]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #57096 [ run ] triggered by Bot. Commit: a2b6e66 Link to invocation

[tensorrt-cicd]

PR_Github #57096 [ run ] completed with state SUCCESS. Commit: a2b6e66
/LLM/main/L0_MergeRequest_PR pipeline #45885 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

[shuyixiong]

/bot run --disable-fail-fast

[shuyixiong]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #57202 [ run ] triggered by Bot. Commit: a2b6e66 Link to invocation

[tensorrt-cicd]

PR_Github #57202 [ run ] completed with state SUCCESS. Commit: a2b6e66
/LLM/main/L0_MergeRequest_PR pipeline #45974 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

[shuyixiong]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #57383 [ run ] triggered by Bot. Commit: a2b6e66 Link to invocation

[tensorrt-cicd]

PR_Github #57383 [ run ] completed with state SUCCESS. Commit: a2b6e66
/LLM/main/L0_MergeRequest_PR pipeline #46132 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