[None][chore] Moving kimi-k2-thinking deployment guide configs to config files.

[fsaady]

Summary by CodeRabbit

• Documentation

  • Updated Kimi K2 Thinking deployment guide with YAML-based configuration approach
  • Added OpenAI-compatible API usage examples and sample responses
  • Improved performance settings guidance with curated configuration references

• New Features

  • Added recommended performance configurations for Kimi K2 Thinking deployment
  • Added SLURM-based configuration for benchmarking workflow

Description

Moved the hard-coded configs from the deployment guide to examples/configs/curated/kimi-k2-thinking.yaml and examples/wide_ep/slurm_scripts/kimi-k2-thinking.yaml.
This will allow easier maintenance of the configs and easier automated testing in PR #11585. It also keeping kimi-k2-thinking aligned with the rest of the deployment guides.

Test Coverage

This PR only changes the docs, testing for all the curated configs will be added in PR #11585.

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)

• 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

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

Details

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

[coderabbitai]

📝 Walkthrough

Walkthrough

This pull request adds curated YAML configuration files and deployment guidance for the Kimi-K2-Thinking model on TensorRT LLM. Two new configuration files are introduced: one for general LLM API settings and another for SLURM-based disaggregated serving. The deployment documentation is restructured to reference these external configs instead of inline YAML, and the model recipes table is updated to include the new configuration.

Changes

Cohort / File(s)	Summary
Documentation Updates docs/source/deployment-guide/deployment-guide-for-kimi-k2-thinking-on-trtllm.md, docs/source/deployment-guide/index.rst	Replaced inline YAML configurations with references to external curated config files. Updated server launch instructions to use EXTRA_LLM_API_FILE and SLURM_CONFIG_FILE variables. Added new row to Model Recipes table for Kimi-K2-Thinking (NVFP4) with GPU targets and configuration file link. Included sample OpenAI-compatible API query and response examples.
Configuration Files examples/configs/curated/kimi-k2-thinking.yaml, examples/wide_ep/slurm_scripts/kimi-k2-thinking.yaml	Introduced two new YAML configuration files. The curated config specifies general LLM API parameters (max batch size, sequence length, parallelism settings, KV cache, and trust remote code). The SLURM config provides comprehensive deployment settings including SLURM submission parameters, benchmark mode, hardware layout, environment variables, and separate gen/ctx worker configurations.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Description check	✅ Passed	PR description includes required sections (Description, Test Coverage, PR Checklist) with clear explanation of changes, rationale, and test coverage planning.
Title check	✅ Passed	The title clearly and concisely describes the main change: moving kimi-k2-thinking deployment guide configurations to dedicated config files, which aligns with the actual changes across all modified files.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

docs/source/deployment-guide/deployment-guide-for-kimi-k2-thinking-on-trtllm.md (1)

47-56: Nitpick: language: shell on a YAML literalinclude loses YAML syntax highlighting

The included file is YAML; specifying language: shell means the rendered docs highlight everything as shell, including the YAML keys. If readers are copying the YAML block to inspect it, they lose the benefit of proper highlighting. Consider language: yaml — the heredoc framing (cat << EOF … EOF) is added via prepend/append so it doesn't affect how the file content is highlighted.

✏️ Proposed change (applies identically to both literalinclude blocks, lines 47–56 and 104–113)

-language: shell
+language: yaml

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In
`@docs/source/deployment-guide/deployment-guide-for-kimi-k2-thinking-on-trtllm.md`
around lines 47 - 56, The literalinclude blocks currently set "language: shell"
which causes YAML content to be highlighted as shell; update the "language" key
in the literalinclude directives to "yaml" (i.e., change language: shell ->
language: yaml) for the literalinclude blocks so the included
../../../examples/configs/curated/kimi-k2-thinking.yaml is highlighted
correctly; look for the literalinclude directives and the "language" key in the
two blocks around the heredoc prepend/append entries and change both
occurrences.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@examples/wide_ep/slurm_scripts/kimi-k2-thinking.yaml`:
- Around line 52-67: cuda_graph_config.batch_sizes contains values larger than
the configured max_batch_size (128), which are unreachable and waste CUDA graph
memory and compile time; update cuda_graph_config.batch_sizes to only include
values <= max_batch_size (trim/remove 256, 512, 768, 1024, 2048) so the list is
bounded by max_batch_size and no invalid batch sizes are created at startup
(refer to cuda_graph_config.batch_sizes and max_batch_size to locate and fix).

---

Nitpick comments:
In
`@docs/source/deployment-guide/deployment-guide-for-kimi-k2-thinking-on-trtllm.md`:
- Around line 47-56: The literalinclude blocks currently set "language: shell"
which causes YAML content to be highlighted as shell; update the "language" key
in the literalinclude directives to "yaml" (i.e., change language: shell ->
language: yaml) for the literalinclude blocks so the included
../../../examples/configs/curated/kimi-k2-thinking.yaml is highlighted
correctly; look for the literalinclude directives and the "language" key in the
two blocks around the heredoc prepend/append entries and change both
occurrences.

ℹ️ Review info

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between df3484d and 70c7051.

📒 Files selected for processing (4)

• docs/source/deployment-guide/deployment-guide-for-kimi-k2-thinking-on-trtllm.md
• docs/source/deployment-guide/index.rst
• examples/configs/curated/kimi-k2-thinking.yaml
• examples/wide_ep/slurm_scripts/kimi-k2-thinking.yaml

[fsaady]

/bot run

[tensorrt-cicd]

PR_Github #36639 [ run ] triggered by Bot. Commit: c30b888 Link to invocation

[tensorrt-cicd]

PR_Github #36639 [ run ] completed with state SUCCESS. Commit: c30b888
/LLM/main/L0_MergeRequest_PR pipeline #28363 completed with status: 'SUCCESS'

Link to invocation