[TRTLLM-11548][doc] Add Qwen3.5 deployment guide doc

[nv-guomingz]

Summary by CodeRabbit

Release Notes

• New Features

  • Added support for Qwen3.5-397B-A17B-NVFP4 model variant with curated configuration for maximum throughput on B200/B300 and GB200/GB300 GPUs.

• Documentation

  • Updated deployment guide with Qwen3.5-specific instructions, GPU requirements, tensor parallel settings, and troubleshooting guidance.
  • Updated supported models table to reflect Qwen3.5 model status.
  • Added example benchmarking configuration for the new model variant.

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 introduces support for the nvidia/Qwen3.5-397B-A17B-NVFP4 model to TensorRT LLM by adding a curated configuration, metadata mappings, and comprehensive deployment documentation. All necessary runtime settings, GPU compatibility declarations, and user-facing guidance are included.

Changes

Qwen3.5 Model Onboarding

Layer / File(s)	Summary
Curated Configuration Setup examples/configs/curated/lookup.yaml, examples/configs/curated/qwen3.5.yaml, scripts/generate_config_table.py, docs/source/_static/config_db.json	Model nvidia/Qwen3.5-397B-A17B-NVFP4 is registered in the model class mapping to Qwen3_5MoeForCausalLM with the new qwen3.5.yaml configuration file, GPU compatibility list B200, B300, GB200, GB300, and "Max Throughput" scenario. The configuration file defines batch/token limits, tensor and MoE parallelism, attention DP, KV cache memory tuning, and postprocess worker settings. Model metadata (display name and Hugging Face URL) is added to config_db.json curated entries and models map, and to the MODEL_INFO table for doc generation.
Deployment Guide and Documentation docs/source/deployment-guide/deployment-guide-for-qwen3.5-on-trtllm.md, docs/source/deployment-guide/index.rst, docs/source/models/supported-models.md	Deployment guide is updated from Qwen3-Next to Qwen3.5, documenting the NVFP4 checkpoint architecture, GPU requirements, Docker container setup referencing qwen3.5.yaml, trtllm-serve launch with qwen3_5/qwen3 parsers, YAML API options including new enable_attention_dp, KV cache guidance, chat-completions request example, and Qwen3.5-specific troubleshooting including reasoning toggle via chat_template_kwargs.enable_thinking. Benchmarking wrapper is updated to use the Qwen3.5 model. Navigation index updated to reference the qwen3.5 guide, and supported-models table footnote removed from Qwen3_5MoeForCausalLM entry.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#14964: Adds "Max Throughput" NVFP4 model documentation for a different model ID using the same curated configuration system and metadata tables.

Suggested reviewers

• QiJune
• laikhtewari

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is incomplete, containing only the template structure without substantive content explaining the changes, objectives, or test coverage.	Fill in the Description section explaining what Qwen3.5 deployment guide was added and why; detail test coverage for documentation changes; ensure the PR title follows the specified format with ticket ID and type.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: adding a Qwen3.5 deployment guide documentation. It includes the JIRA ticket, type indicator, and concise description.
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.}

[nv-guomingz]

/bot run

[tensorrt-cicd]

PR_Github #52802 [ run ] triggered by Bot. Commit: 4f8cdd5 Link to invocation

[arysef]

approved with suggestions

[nv-guomingz]

/bot run

[tensorrt-cicd]

PR_Github #52815 [ run ] triggered by Bot. Commit: 23d4cbf Link to invocation

[tensorrt-cicd]

PR_Github #52802 [ run ] completed with state ABORTED. Commit: 4f8cdd5

Link to invocation

[tensorrt-cicd]

PR_Github #52815 [ run ] completed with state SUCCESS. Commit: 23d4cbf
/LLM/main/L0_MergeRequest_PR pipeline #42069 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

[nv-guomingz]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #52881 [ run ] triggered by Bot. Commit: 23d4cbf Link to invocation

[tensorrt-cicd]

PR_Github #52881 [ run ] completed with state SUCCESS. Commit: 23d4cbf
/LLM/main/L0_MergeRequest_PR pipeline #42132 completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

CI Report

Link to invocation