[#9656][feat] Load default sampling parameters (repetition_penalty, temperature, top_p, top_k and min_p) from generation_config.json #10329
Conversation
|
/bot run |
📝 WalkthroughWalkthroughThe changes introduce a default sampling parameters mechanism and update the OpenAI protocol layer to support optional sampling configuration fields. A new constant defines defaults for repetition_penalty, temperature, top_p, top_k, and min_p, which are loaded during SamplingParams initialization with fallback support. API request models now accept these parameters as Optional fields instead of required numeric defaults. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Pre-merge checks and finishing touches❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
✨ Finishing touches
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
tensorrt_llm/sampling_params.py (1)
123-202: Update docstring to document default parameter loading behavior.The class docstring should mention that default values for
repetition_penalty,temperature,top_p,top_k, andmin_pcan be loaded from the model'sgeneration_config.jsonwhen not explicitly set by the user.📝 Suggested docstring addition
Add the following to the class docstring after the existing parameter descriptions:
spaces_between_special_tokens (bool): Whether to add spaces between special tokens in the output. Defaults to True. + + Note: + If not explicitly set, the following parameters will be loaded from the model's generation_config.json: + repetition_penalty, temperature, top_p, top_k, and min_p. If unavailable in generation_config, + they default to the values defined in DEFAULT_SAMPLING_PARAMS. """ # noqa: E501
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
tensorrt_llm/sampling_params.pytensorrt_llm/serve/openai_protocol.py
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Code developed for TensorRT-LLM should conform to Python 3.8+
Indent Python code with 4 spaces. Do not use tabs
Always maintain the namespace when importing in Python, even if only one class or function from a module is used
Python files should use snake_case naming:some_file.py
Python classes should use PascalCase naming:class SomeClass
Python functions and methods should use snake_case naming:def my_awesome_function():
Python local variables should use snake_case naming:my_variable = ...
Python variable names that start with a number should be prefixed with 'k':k_99th_percentile = ...
Python global variables should use upper snake_case with prefix 'G':G_MY_GLOBAL = ...
Python constants should use upper snake_case naming:MY_CONSTANT = ...
Avoid shadowing variables declared in an outer scope in Python
Initialize all externally visible members of a Python class in the constructor
For Python interfaces that may be used outside a file, prefer docstrings over comments
Python comments should be reserved for code within a function, or interfaces that are local to a file
Use Google style docstrings in Python for classes and functions, which can be parsed by Sphinx
Python attributes and variables can be documented inline with type and description
Avoid using reflection in Python when functionality can be easily achieved without reflection
When using try-except blocks in Python, limit the except to the smallest set of errors possible
When using try-except blocks in Python to handle multiple possible variable types (duck-typing), keep the body of the try as small as possible, using the else block for logic
Files:
tensorrt_llm/sampling_params.pytensorrt_llm/serve/openai_protocol.py
**/*.{cpp,h,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the year of its latest meaningful modification
Files:
tensorrt_llm/sampling_params.pytensorrt_llm/serve/openai_protocol.py
🧠 Learnings (2)
📓 Common learnings
Learnt from: MatthiasKohl
Repo: NVIDIA/TensorRT-LLM PR: 6904
File: cpp/tensorrt_llm/pybind/thop/bindings.cpp:55-57
Timestamp: 2025-08-14T15:38:01.771Z
Learning: In TensorRT-LLM Python bindings, tensor parameter collections like mla_tensor_params and spec_decoding_tensor_params are kept as required parameters without defaults to maintain API consistency, even when it might affect backward compatibility.
📚 Learning: 2025-08-14T15:38:01.771Z
Learnt from: MatthiasKohl
Repo: NVIDIA/TensorRT-LLM PR: 6904
File: cpp/tensorrt_llm/pybind/thop/bindings.cpp:55-57
Timestamp: 2025-08-14T15:38:01.771Z
Learning: In TensorRT-LLM Python bindings, tensor parameter collections like mla_tensor_params and spec_decoding_tensor_params are kept as required parameters without defaults to maintain API consistency, even when it might affect backward compatibility.
Applied to files:
tensorrt_llm/sampling_params.pytensorrt_llm/serve/openai_protocol.py
🧬 Code graph analysis (1)
tensorrt_llm/serve/openai_protocol.py (1)
tensorrt_llm/_torch/pyexecutor/model_engine.py (1)
use_beam_search(543-544)
🔇 Additional comments (5)
tensorrt_llm/serve/openai_protocol.py (4)
256-257: LGTM! Sampling parameters now support optional values.These fields are correctly changed to
OptionalwithNonedefaults, allowing the system to load default values fromgeneration_config.jsoninSamplingParams._setuprather than hardcoding them at the protocol layer. This aligns with the PR objective to match behavior with vLLM and SGLang.Also applies to: 264-264, 266-267
312-313: LGTM! Parameters correctly passed without fallback logic.The temperature and top_p parameters are now passed as-is (potentially
None) toSamplingParams, where defaults will be applied in the_setupmethod. This correctly implements the new default-loading mechanism.
554-555: LGTM! Consistent with CompletionRequest changes.These
ChatCompletionRequestfields mirror the changes inCompletionRequest, ensuring consistent behavior across both API endpoints for loading default sampling parameters.Also applies to: 573-573, 575-576
659-659: LGTM! Consistent parameter passing.Parameters are correctly passed without fallback logic, consistent with the
CompletionRequest.to_sampling_paramsimplementation.Also applies to: 666-666
tensorrt_llm/sampling_params.py (1)
112-118: LGTM! Well-defined default sampling parameters.The constant follows naming conventions and provides sensible defaults that represent "no filtering" states for sampling parameters. These values align with standard practice in LLM inference (e.g.,
top_k=0meaning "all logits" per the docstring at line 153).
Signed-off-by: Dmitry Barsukoff <riZZZhik@gmail.com>
riZZZhik
force-pushed
the
add_load-sampling-params-from-generation-config
branch
from
12cd90e to
2862587
Compare
svc-trtllm-gh-bot
added
the
Community want to contribute
Signed-off-by: Dmitry Barsukoff <riZZZhik@gmail.com>
riZZZhik
force-pushed
the
add_load-sampling-params-from-generation-config
branch
3 times, most recently
from
7efb93c to
dde19da
Compare
|
/bot run |
2 participants
Description
Modern hf-like checkpoints include default sampling parameters (repetition_penalty, temperature, top_p, top_k, and min_p) in
generation_config.json, but TensorRT-LLM does not load them as defaults, unlike vLLM and SGLang.This PR adds support for loading these parameters.
More info in #9656
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)
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 thestage-listparameter 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.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip 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-pipelineReuse 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.