[None][feat] Cherry-pick Responses API and multiple postprocess workers support for chat harmony #7600
Conversation
…NVIDIA#7341) Signed-off-by: Junyi Xu <[email protected]>
…pletions API (NVIDIA#7508) Signed-off-by: Junyi Xu <[email protected]> Co-authored-by: Pengyun Lin <[email protected]>
Signed-off-by: Junyi Xu <[email protected]>
JunyiXu-nv
force-pushed
the
release/1.1.0rc2
branch
from
ca68f6c to
b38bd4a
Compare
|
/bot run |
|
Caution Review failedFailed to post review comments. Configuration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro 📒 Files selected for processing (9)
🧰 Additional context used📓 Path-based instructions (3)**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Files:
**/*.py📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Files:
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Files:
🧠 Learnings (1)📚 Learning: 2025-07-28T17:06:08.621ZApplied to files:
🧬 Code graph analysis (7)tests/integration/defs/test_e2e.py (1)
tensorrt_llm/serve/responses_utils.py (3)
tensorrt_llm/serve/openai_server.py (5)
tests/unittest/llmapi/apps/_test_openai_responses.py (1)
tensorrt_llm/serve/harmony_adapter.py (3)
tensorrt_llm/serve/postprocess_handlers.py (3)
tensorrt_llm/serve/openai_protocol.py (1)
🪛 Ruff (0.12.2)tests/integration/defs/test_e2e.py1516-1516: Unused function argument: (ARG001) tensorrt_llm/serve/responses_utils.py88-88: f-string without any placeholders Remove extraneous (F541) 122-122: Avoid specifying long messages outside the exception class (TRY003) 128-128: Use of (S101) 137-137: Avoid specifying long messages outside the exception class (TRY003) 163-163: Do not use mutable data structures for argument defaults Replace with (B006) 196-196: f-string without any placeholders Remove extraneous (F541) 219-219: f-string without any placeholders Remove extraneous (F541) 224-224: Use of (S101) 236-236: f-string without any placeholders Remove extraneous (F541) 322-322: Avoid specifying long messages outside the exception class (TRY003) 344-344: Do not use mutable data structures for argument defaults Replace with (B006) 408-408: Avoid specifying long messages outside the exception class (TRY003) 423-423: Avoid specifying long messages outside the exception class (TRY003) 468-468: Avoid specifying long messages outside the exception class (TRY003) 488-488: Avoid specifying long messages outside the exception class (TRY003) 503-503: Avoid specifying long messages outside the exception class (TRY003) 513-513: Value being cast to Remove unnecessary (RUF046) 529-529: f-string without any placeholders Remove extraneous (F541) 559-559: PEP 484 prohibits implicit Convert to (RUF013) 573-573: Avoid specifying long messages outside the exception class (TRY003) 618-618: PEP 484 prohibits implicit Convert to (RUF013) 673-673: Use of (S101) tensorrt_llm/serve/openai_server.py803-803: Store a reference to the return value of (RUF006) 815-815: Do not catch blind exception: (BLE001) 835-835: Abstract (TRY301) 867-867: Store a reference to the return value of (RUF006) 887-887: Do not catch blind exception: (BLE001) tests/unittest/llmapi/apps/_test_openai_responses.py40-40: Use of (S101) 41-41: Use of (S101) 57-57: Use of (S101) 58-58: Use of (S101) 62-62: Use of (S101) 63-63: Use of (S101) 121-121: Unused function argument: (ARG001) 156-156: Use of (S101) 190-190: Use of (S101) 191-191: Use of (S101) 237-237: Use of (S101) 238-238: Use of (S101) 239-239: Use of (S101) tensorrt_llm/serve/harmony_adapter.py157-157: Loop control variable Rename unused (B007) 1287-1287: Consider moving this statement to an (TRY300) 1574-1574: Consider moving this statement to an (TRY300) 1581-1581: Use Remove exception name (TRY201) tensorrt_llm/serve/openai_protocol.py718-718: Use explicit conversion flag Replace with conversion flag (RUF010) 774-774: Avoid specifying long messages outside the exception class (TRY003) 781-781: Avoid specifying long messages outside the exception class (TRY003) 802-802: Use explicit conversion flag Replace with conversion flag (RUF010) ⏰ 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)
📝 WalkthroughWalkthroughIntroduces a stateful Harmony adapter singleton, remodels streaming into batch returns, adds tool-aware processing and usage computation, and implements an OpenAI-compatible Responses API (request model, response models, streaming events, persistence). Refactors server routing to expose /v1/responses, adds post-processing handlers, a conversation history store, and comprehensive unit/integration tests. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Client
participant OpenAIServer as OpenAIServer (/v1/responses)
participant ResponsesUtils as responses_utils
participant Harmony as HarmonyAdapter (singleton)
Client->>OpenAIServer: POST /v1/responses (ResponsesRequest)
OpenAIServer->>ResponsesUtils: request_preprocess(request, prev_response, adapter, store, enable_store)
ResponsesUtils->>Harmony: construct_harmony_messages + render_for_completion
ResponsesUtils-->>OpenAIServer: sampling_params, inputs, prev_msgs
OpenAIServer->>Harmony: generate(inputs, sampling_params, _postproc_params)
Harmony-->>OpenAIServer: generation_result (non-streaming)
OpenAIServer->>ResponsesUtils: create_response(generator=None, request, sampling_params, model, store, generation_result)
ResponsesUtils-->>OpenAIServer: ResponsesResponse
OpenAIServer-->>Client: 200 OK JSON (ResponsesResponse)
note over OpenAIServer,Harmony: Non-streaming flow
sequenceDiagram
autonumber
participant Client
participant OpenAIServer as OpenAIServer (/v1/responses?stream=true)
participant ResponsesUtils as responses_utils
participant Harmony as HarmonyAdapter (singleton)
participant SSE as SSE Stream
Client->>OpenAIServer: POST /v1/responses (stream=true)
OpenAIServer->>ResponsesUtils: request_preprocess(...)
ResponsesUtils-->>OpenAIServer: sampling_params, inputs
OpenAIServer->>Harmony: create_streaming_generator(inputs, sampling_params, _postproc_params)
OpenAIServer->>ResponsesUtils: process_streaming_events(request, sampling_params, generator, adapter, model, store)
loop streaming batches
ResponsesUtils-->>SSE: event: response.created / in_progress / deltas
Harmony-->>ResponsesUtils: token batches, finish_reason
end
ResponsesUtils-->>SSE: event: response.completed (+final, usage)
SSE-->>Client: SSE stream ends
note over ResponsesUtils: Emits reasoning/text/tool call deltas and final payload
Estimated code review effort🎯 5 (Critical) | ⏱️ ~120+ minutes Possibly related PRs
Suggested labels
Suggested reviewers
✨ Finishing Touches
🧪 Generate unit tests
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 #17975 [ run ] triggered by Bot |
|
PR_Github #17975 [ run ] completed with state |
|
/bot run |
|
PR_Github #17995 [ run ] triggered by Bot |
|
PR_Github #17995 [ run ] completed with state |
|
/bot run |
|
PR_Github #18021 [ run ] triggered by Bot |
|
PR_Github #18021 [ run ] completed with state |
|
/bot run |
|
PR_Github #18072 [ run ] triggered by Bot |
|
PR_Github #18072 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #18109 [ run ] triggered by Bot |
litaotju
added
the
Release Blocker
|
PR_Github #18109 [ run ] completed with state |
|
The failure is not related to this change. Merging. |
Summary by CodeRabbit
New Features
Refactor
Tests
Description
This PR includes:
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.
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.