[TRTLLM-9522][test] cover LLM API multi_modal_embeddings
#9963
Conversation
|
/bot run |
|
PR_Github #28070 [ run ] triggered by Bot. Commit: |
|
PR_Github #28070 [ run ] completed with state |
|
/bot run |
|
PR_Github #28073 [ run ] triggered by Bot. Commit: |
|
PR_Github #28073 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #28116 [ run ] triggered by Bot. Commit: |
|
PR_Github #28116 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #28139 [ run ] triggered by Bot. Commit: |
|
PR_Github #28139 [ run ] completed with state |
ixlmar
force-pushed
the
test/mm-embeddings
branch
from
94d9939 to
414b584
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #30896 [ run ] triggered by Bot. Commit: |
|
PR_Github #30896 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #31028 [ run ] triggered by Bot. Commit: |
|
/bot kill |
ixlmar
force-pushed
the
test/mm-embeddings
branch
from
414b584 to
7856281
Compare
📝 WalkthroughWalkthroughThis pull request restructures the multimodal encoder test suite by introducing KV-cache event validation, parameterized fixtures across multiple models (LLAVA, Qwen2.5 VL, Qwen3 VL), encoder-centered test paths, and embedding round-trip handling in batch scenarios. The changes span 269 additions and 210 removals within a single test file. Changes
Estimated code review effort🎯 4 (Complex) | ⏱️ ~50 minutes 🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ 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
🤖 Fix all issues with AI agents
In @tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py:
- Around line 1-6: Add the required NVIDIA copyright header to the top of the
test file test_mm_encoder_standalone.py: insert the standard multi-line NVIDIA
copyright comment block including the year of latest meaningful modification and
the canonical NVIDIA wording before any imports so the file conforms to the
project's licensing/coding guidelines.
🧹 Nitpick comments (4)
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py (4)
82-83: Potential test flakiness with fixed sleep.Using a fixed
time.sleep(0.5)to wait for events to be dispatched can cause intermittent failures under load or on slower systems. Consider polling with a timeout or using an event/condition mechanism if available.♻️ Possible polling approach
- time.sleep(0.5) # Wait for events to be dispatched - events = llm.get_kv_cache_events(10) + # Poll for events with timeout instead of fixed sleep + max_wait_time = 2.0 + poll_interval = 0.1 + elapsed = 0.0 + events = [] + while elapsed < max_wait_time: + events = llm.get_kv_cache_events(10) + if events: + break + time.sleep(poll_interval) + elapsed += poll_interval
114-116: Type hints use Python 3.10+ syntax; consider 3.8+ compatibility.The type hints
tuple[LLM, LLM | None]andGenerator[tuple[...], None, None]use Python 3.10+ syntax (PEP 604 union operator and lowercase generics). As per coding guidelines, TensorRT-LLM should conform to Python 3.8+.♻️ Python 3.8+ compatible type hints
-from typing import Generator +from typing import Generator, Optional, Tuple ... @pytest.fixture(scope="module") def llms(model_dir: Path, - pd_disagg: bool) -> Generator[tuple[LLM, LLM | None], None, None]: + pd_disagg: bool) -> Generator[Tuple[LLM, Optional[LLM]], None, None]:Alternatively, add
from __future__ import annotationsat the top of the file to enable postponed evaluation of annotations.
343-348: Consider addingstrict=Truetozip()calls for defensive validation.While the lengths are validated elsewhere, adding
strict=Truetozip()calls would catch length mismatches earlier and make the intent explicit. This applies to multiplezip()calls in this function (lines 343, 361, 400, 405).♻️ Example fix for line 343
- for input, encoder_output in zip(inputs, encoder_outputs): + for input, encoder_output in zip(inputs, encoder_outputs, strict=True):Apply similar changes to other
zip()calls in this function.
377-380: Clarify the single-element unpacking intent.The comma unpacking
mm_embed, = ...assumes exactly one element. While this is valid, a brief comment would clarify the expectation.- mm_embed, = input_with_embedding["multi_modal_embeddings"][ - "image"] + # Expect exactly one embedding tensor per image + (mm_embed,) = input_with_embedding["multi_modal_embeddings"][ + "image"]
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: The 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 Python modules, even if only one class or function from a module is used
Python filenames should use snake_case (e.g.,some_file.py)
Python classes should use PascalCase (e.g.,class SomeClass)
Python functions and methods should use snake_case (e.g.,def my_awesome_function():)
Python local variables should use snake_case, with prefixkfor variable names that start with a number (e.g.,k_99th_percentile)
Python global variables should use upper snake_case with prefixG(e.g.,G_MY_GLOBAL)
Python constants should use upper snake_case (e.g.,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
Use comments in Python for code within a function, or interfaces that are local to a file
Use Google-style docstrings for Python classes and functions, which can be parsed by Sphinx
Python attributes and variables can be documented inline with the format"""<type>: Description"""
Avoid using reflection in Python when functionality can be easily achieved without reflection
When using try-except blocks in Python, limit the except clause 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 and use the else block for the main logic
Files:
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py
**/*.{cpp,cc,cxx,h,hpp,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM source files (.cpp, .h, .cu, .py, and other source files) should contain an NVIDIA copyright header with the year of latest meaningful modification
Files:
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py
🧠 Learnings (2)
📓 Common learnings
Learnt from: moraxu
Repo: NVIDIA/TensorRT-LLM PR: 6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 7645
File: tests/integration/test_lists/qa/llm_function_core.txt:648-648
Timestamp: 2025-09-09T09:40:45.658Z
Learning: In TensorRT-LLM test lists, it's common and intentional for the same test to appear in multiple test list files when they serve different purposes (e.g., llm_function_core.txt for comprehensive core functionality testing and llm_function_core_sanity.txt for quick sanity checks). This duplication allows tests to be run in different testing contexts.
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
Repo: NVIDIA/TensorRT-LLM PR: 6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py
🧬 Code graph analysis (1)
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py (2)
tensorrt_llm/llmapi/mm_encoder.py (2)
MultimodalEncoder(16-136)generate(80-118)tensorrt_llm/llmapi/llm.py (2)
LLM(1171-1187)generate(266-348)
🪛 Ruff (0.14.10)
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py
343-343: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
361-362: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
400-401: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
405-405: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
⏰ 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 (3)
tests/unittest/_torch/multimodal/test_mm_encoder_standalone.py (3)
151-173: LGTM!The helper function is well-structured with appropriate assertions for input validation.
176-247: LGTM!The test properly validates that encoder-assisted generation produces identical results to standard generation. The disaggregated mode handling with separate decode LLM is correctly implemented.
286-314: Good use of skip conditions for unsupported configurations.The skip logic correctly handles:
- Qwen models lacking
attach_multimodal_embeddings- Disaggregated mode not implemented for batch tests
- Redundant test configurations
|
PR_Github #31028 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #31088 [ run ] triggered by Bot. Commit: |
|
PR_Github #31088 [ run ] completed with state |
|
/bot run --disable-fail-fast |
Signed-off-by: ixlmar <[email protected]>
ixlmar
force-pushed
the
test/mm-embeddings
branch
from
7856281 to
589ca8a
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #31222 [ run ] triggered by Bot. Commit: |
|
PR_Github #31223 [ run ] triggered by Bot. Commit: |
|
PR_Github #31223 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #31272 [ run ] triggered by Bot. Commit: |
|
PR_Github #31272 [ run ] completed with state |
2 participants
Description
This improves test coverage for pre-existing functionality affected by #9715.
Note: Consider reviewing with something like
git diff -w --color-moved --color-moved-ws=no.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.
Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.