[None][feat] Support Qwen3 reasoning parser #8000
Conversation
LinPoly
changed the title
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
b8b9617 to
f78ec8c
Compare
|
/bot run |
📝 WalkthroughWalkthroughRemoves the public boolean Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Model as Model Output (tokens)
participant Post as postprocess_handlers.apply_reasoning_parser
participant Parser as DeepSeekR1Parser / Qwen3
participant Client as Client Stream
Model-->>Post: delta_text
Post->>Parser: parse_delta(delta_text)
alt waiting for start/end tags
Parser-->>Post: ReasoningParserResult(content="", reasoning_content="")
Note right of Parser #f9f871: buffer/state updated
else yields reasoning/content
Parser-->>Post: ReasoningParserResult(content=..., reasoning_content=...)
Post-->>Client: DeltaMessage(content, reasoning_content)
end
sequenceDiagram
autonumber
participant Input as Full Text
participant Parser as DeepSeekR1Parser
participant Caller as Caller
Caller->>Parser: parse(text)
alt reasoning_at_start = True
Parser->>Parser: partition by think tags (start at beginning)
else reasoning_at_start = False
Parser->>Parser: drop text before first start tag
end
Parser-->>Caller: ReasoningParserResult(content=..., reasoning_content=...)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests
Tip 👮 Agentic pre-merge checks are now available in preview!Pro plan users can now enable pre-merge checks in their settings to enforce checklists before merging PRs.
Please see the documentation for more information. Example: reviews:
pre_merge_checks:
custom_checks:
- name: "Undocumented Breaking Changes"
mode: "warning"
instructions: |
Pass/fail criteria: All breaking changes to public APIs, CLI flags, environment variables, configuration keys, database schemas, or HTTP/GraphQL endpoints must be documented in the "Breaking Change" section of the PR description and in CHANGELOG.md. Exclude purely internal or private changes (e.g., code not exported from package entry points or explicitly marked as internal).Please share your feedback with us on this Discord post. 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 |
|
PR_Github #19963 [ run ] triggered by Bot |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
tensorrt_llm/llmapi/reasoning_parser.py (1)
107-121: Fix parser registry typing and case-insensitive lookup (breaks Py3.8 and mixed-case inputs).
- dict[...] and type[...] are Python 3.9+; this file must target Python 3.8+. Use typing.Dict/Type.
- Membership check uses raw string, but retrieval uses .lower(). Mixed-case inputs (e.g., "DeepSeek-R1") will incorrectly raise ValueError.
Apply this diff:
- parsers: dict[str, BaseReasoningParser] = { - "deepseek-r1": DeepSeekR1Parser, - "qwen3": DeepSeekR1Parser, - } + parsers: Dict[str, Type[BaseReasoningParser]] = { + "deepseek-r1": DeepSeekR1Parser, + "qwen3": DeepSeekR1Parser, + } @@ - def create_reasoning_parser(reasoning_parser: str) -> BaseReasoningParser: - if reasoning_parser not in ReasoningParserFactory.parsers: - raise ValueError(f"Invalid reasoning_parser: {reasoning_parser}") - reasoning_parser_class = ReasoningParserFactory.parsers.get( - reasoning_parser.lower()) - if reasoning_parser == "deepseek-r1": - return reasoning_parser_class(reasoning_at_start=True) - return reasoning_parser_class() + def create_reasoning_parser(reasoning_parser: str) -> BaseReasoningParser: + key = reasoning_parser.lower() + if key not in ReasoningParserFactory.parsers: + raise ValueError(f"Invalid reasoning_parser: {reasoning_parser}") + reasoning_parser_class = ReasoningParserFactory.parsers[key] + if key == "deepseek-r1": + return reasoning_parser_class(reasoning_at_start=True) + return reasoning_parser_class()Add these imports near the top of the file:
from typing import Dict, Typetensorrt_llm/serve/postprocess_handlers.py (1)
1-1: Add NVIDIA Apache-2.0 header.Required by coding guidelines.
🧹 Nitpick comments (6)
tensorrt_llm/llmapi/reasoning_parser.py (2)
37-47: Remove unused helper (_create_reasoning_end_result).Dead code; not referenced.
Apply this diff:
- def _create_reasoning_end_result(self, content: str, - reasoning_content: str): - if len(content) == 0: - reasoning_parser_result = ReasoningParserResult( - reasoning_content=reasoning_content) - elif len(reasoning_content) == 0: - reasoning_parser_result = ReasoningParserResult(content=content) - else: - reasoning_parser_result = ReasoningParserResult( - content=content, reasoning_content=reasoning_content) - return reasoning_parser_result
103-103: Tighten unreachable-path raise to satisfy TRY003.Use a shorter message or AssertionError.
Apply this diff:
- raise RuntimeError("Unreachable code reached.") + raise AssertionError("unreachable")tensorrt_llm/serve/postprocess_handlers.py (1)
172-179: Prefer isinstance over type(...) is ...This avoids surprises with subclasses and is the idiomatic check.
Apply this diff:
- if args.tool_choice and type( - args.tool_choice) is ChatCompletionNamedToolChoiceParam: + if args.tool_choice and isinstance( + args.tool_choice, ChatCompletionNamedToolChoiceParam):tests/unittest/llmapi/apps/_test_openai_reasoning.py (2)
83-89: Use model_dump for Pydantic v2 forward-compat.delta.dict() is deprecated in Pydantic v2. Use model_dump() instead.
Apply this diff:
- delta = choice.delta.dict() + delta = choice.delta.model_dump()
1-1: Add NVIDIA Apache-2.0 header to test file.Aligns with repository guidelines.
tests/unittest/llmapi/test_reasoning_parser.py (1)
1-1: Add NVIDIA Apache-2.0 header to test file.Per coding guidelines.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tensorrt_llm/llmapi/reasoning_parser.py(3 hunks)tensorrt_llm/serve/postprocess_handlers.py(6 hunks)tests/unittest/llmapi/apps/_test_openai_reasoning.py(5 hunks)tests/unittest/llmapi/test_reasoning_parser.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
tests/unittest/llmapi/test_reasoning_parser.pytensorrt_llm/serve/postprocess_handlers.pytensorrt_llm/llmapi/reasoning_parser.pytests/unittest/llmapi/apps/_test_openai_reasoning.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.
Files:
tests/unittest/llmapi/test_reasoning_parser.pytensorrt_llm/serve/postprocess_handlers.pytensorrt_llm/llmapi/reasoning_parser.pytests/unittest/llmapi/apps/_test_openai_reasoning.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
tests/unittest/llmapi/test_reasoning_parser.pytensorrt_llm/serve/postprocess_handlers.pytensorrt_llm/llmapi/reasoning_parser.pytests/unittest/llmapi/apps/_test_openai_reasoning.py
🧬 Code graph analysis (3)
tests/unittest/llmapi/test_reasoning_parser.py (1)
tensorrt_llm/llmapi/reasoning_parser.py (6)
ReasoningParserFactory(106-120)create_reasoning_parser(113-120)parse(14-15)parse(49-61)parse_delta(18-19)parse_delta(63-103)
tensorrt_llm/serve/postprocess_handlers.py (1)
tensorrt_llm/serve/openai_protocol.py (1)
DeltaMessage(445-451)
tests/unittest/llmapi/apps/_test_openai_reasoning.py (2)
tests/integration/defs/stress_test/stress_test.py (2)
model_name(110-112)get_model_path(310-312)tests/unittest/llmapi/apps/_test_openai_misc.py (2)
model_name(19-26)backend(14-15)
🪛 Ruff (0.13.1)
tensorrt_llm/llmapi/reasoning_parser.py
103-103: Avoid specifying long messages outside the exception class
(TRY003)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (4)
tensorrt_llm/serve/postprocess_handlers.py (1)
112-117: API shape update looks good.Returning (content, reasoning_content) and defaulting to ("text", "") when parser is absent is clear and compatible with downstream usage.
tests/unittest/llmapi/apps/_test_openai_reasoning.py (1)
120-127: Potential flakiness: zero content chunks assumption.Asserting len(content_chunks) == 0 with max_completion_tokens=2 may vary across templates/tokenizers. Consider relaxing or gating this with model-specific conditions.
tests/unittest/llmapi/test_reasoning_parser.py (2)
43-56: qwen3 parsing tests align with new (content, reasoning_content) contract.Good coverage of edge cases (no end tag, only start tag).
58-72: Streaming delta tests look solid.They exercise partial-tag buffering and boundary conditions well.
|
PR_Github #19963 [ run ] completed with state |
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
f78ec8c to
635ac72
Compare
|
/bot run |
|
PR_Github #20337 [ run ] triggered by Bot |
|
PR_Github #20337 [ run ] completed with state |
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
e7e9b94 to
f1930f9
Compare
|
/bot run |
|
PR_Github #20871 [ run ] triggered by Bot |
|
PR_Github #20871 [ run ] completed with state |
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
f1930f9 to
551f9ea
Compare
|
/bot run |
|
PR_Github #21077 [ run ] triggered by Bot |
|
PR_Github #21077 [ run ] completed with state |
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
551f9ea to
8c64ec2
Compare
|
/bot run |
|
PR_Github #21148 [ run ] triggered by Bot |
|
PR_Github #21148 [ run ] completed with state |
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
LinPoly
force-pushed
the
pengyunl/qwen_reasoning_parsing
branch
from
8c64ec2 to
30b50f5
Compare
|
/bot run |
|
PR_Github #21445 [ run ] triggered by Bot |
|
PR_Github #21445 [ run ] completed with state |
|
/bot run |
|
PR_Github #21530 [ run ] triggered by Bot |
|
PR_Github #21530 [ run ] completed with state |
|
/bot run |
|
PR_Github #21579 [ run ] triggered by Bot |
|
PR_Github #21579 [ run ] completed with state |
|
/bot run |
|
PR_Github #21670 [ run ] triggered by Bot |
|
PR_Github #21670 [ run ] completed with state |
|
/bot run |
|
PR_Github #21854 [ run ] triggered by Bot. Commit: |
|
PR_Github #21854 [ run ] completed with state |
|
/bot run |
|
PR_Github #21919 [ run ] triggered by Bot. Commit: |
|
PR_Github #21919 [ run ] completed with state |
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com> Signed-off-by: yufeiwu-nv <230315618+yufeiwu-nv@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com>
4 participants
Summary by CodeRabbit
New Features
Refactor
Tests
Documentation
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)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
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.