[#10707][fix] AutoDeploy: Super accuracy test fixes#10717
[#10707][fix] AutoDeploy: Super accuracy test fixes#10717galagam merged 5 commits intoNVIDIA:mainfrom
Conversation
📝 WalkthroughWalkthroughThe changes introduce FP8 and FP4 quantization test pathways for the Nemotron-Super-V3 model, add corresponding test methods with quantization configurations, update accuracy reference values, and modify integration test lists to execute the new quantization tests on H100 and B200 GPUs. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 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: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
tests/integration/defs/accuracy/test_llm_api_autodeploy.py (2)
17-17: Duplicate import ofllm_models_rootwill shadow the first import.Line 17 imports
llm_models_rootfromtest_common.llm_data, but line 23 re-imports it from..conftest. The second import shadows the first. Based on existing usage in this file (e.g., lines 86, 143-144, 229-231), the import fromtest_common.llm_dataappears to be the intended one.Proposed fix
-from ..conftest import get_device_count, llm_models_root +from ..conftest import get_device_countAlso applies to: 23-23
264-277: Testtest_bf16is not parametrized but test lists referencetest_bf16[4].The test lists (
l0_dgx_h100.ymlandl0_dgx_b200.yml) referencetest_bf16[4], which implies this test should be parametrized withworld_size. However, the method currently hardcodesworld_size=4without using@pytest.mark.parametrize.This mismatch will cause test discovery to fail since pytest expects a parametrized test ID
[4].Proposed fix: Add parametrization to match test list expectations
# 180GB works, might be able to go lower `@pytest.mark.skip_less_device_memory`(180000) `@pytest.mark.skip_less_device`(4) - def test_bf16(self): + `@pytest.mark.parametrize`("world_size", [4]) + def test_bf16(self, world_size): kwargs = self.get_default_kwargs() sampling_params = self.get_default_sampling_params() with AutoDeployLLM(model=self.MODEL_PATH_BF16, tokenizer=self.MODEL_PATH_BF16, - world_size=4, + world_size=world_size, **kwargs) as llm: task = MMLU(self.MODEL_NAME) task.evaluate(llm, sampling_params=sampling_params) task = GSM8K(self.MODEL_NAME) task.evaluate(llm)
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tests/integration/defs/accuracy/references/mmlu.yamltests/integration/defs/accuracy/test_llm_api_autodeploy.pytests/integration/test_lists/test-db/l0_dgx_b200.ymltests/integration/test_lists/test-db/l0_dgx_h100.yml
🧰 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/integration/defs/accuracy/test_llm_api_autodeploy.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/integration/defs/accuracy/test_llm_api_autodeploy.py
🧠 Learnings (7)
📓 Common learnings
Learnt from: Fridah-nv
Repo: NVIDIA/TensorRT-LLM PR: 7227
File: tests/unittest/_torch/auto_deploy/_utils_test/_model_test_utils.py:269-275
Timestamp: 2025-08-27T16:59:12.325Z
Learning: In FP8 quantized linear layers, bias should be kept in high precision (typically float32) rather than being quantized to FP8 or cast to half precision, as bias is added after the matrix multiplication and high precision bias helps maintain numerical accuracy.
📚 Learning: 2025-09-09T09:40:45.658Z
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.
Applied to files:
tests/integration/test_lists/test-db/l0_dgx_h100.ymltests/integration/test_lists/test-db/l0_dgx_b200.ymltests/integration/defs/accuracy/test_llm_api_autodeploy.py
📚 Learning: 2025-08-26T09:49:04.956Z
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7192
File: tests/integration/test_lists/test-db/l0_dgx_b200.yml:56-72
Timestamp: 2025-08-26T09:49:04.956Z
Learning: In TensorRT-LLM test configuration files, the test scheduling system handles wildcard matching with special rules that prevent duplicate test execution even when the same tests appear in multiple yaml files with overlapping GPU wildcards (e.g., "*b200*" and "*gb200*").
Applied to files:
tests/integration/test_lists/test-db/l0_dgx_h100.ymltests/integration/test_lists/test-db/l0_dgx_b200.ymltests/integration/defs/accuracy/test_llm_api_autodeploy.py
📚 Learning: 2025-09-17T02:48:52.732Z
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7781
File: tests/integration/test_lists/waives.txt:313-313
Timestamp: 2025-09-17T02:48:52.732Z
Learning: In TensorRT-LLM, `tests/integration/test_lists/waives.txt` is specifically for waiving/skipping tests, while other test list files like those in `test-db/` and `qa/` directories are for different test execution contexts (pre-merge, post-merge, QA tests). The same test appearing in both waives.txt and execution list files is intentional - the test is part of test suites but will be skipped due to the waiver.
Applied to files:
tests/integration/test_lists/test-db/l0_dgx_h100.ymltests/integration/test_lists/test-db/l0_dgx_b200.yml
📚 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/integration/test_lists/test-db/l0_dgx_h100.ymltests/integration/test_lists/test-db/l0_dgx_b200.ymltests/integration/defs/accuracy/test_llm_api_autodeploy.py
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
Repo: NVIDIA/TensorRT-LLM PR: 6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
tests/integration/defs/accuracy/test_llm_api_autodeploy.py
📚 Learning: 2025-08-11T20:09:24.389Z
Learnt from: achartier
Repo: NVIDIA/TensorRT-LLM PR: 6763
File: tests/integration/defs/triton_server/conftest.py:16-22
Timestamp: 2025-08-11T20:09:24.389Z
Learning: In the TensorRT-LLM test infrastructure, the team prefers simple, direct solutions (like hard-coding directory traversal counts) over more complex but robust approaches when dealing with stable directory structures. They accept the maintenance cost of updating tests if the layout changes.
Applied to files:
tests/integration/defs/accuracy/test_llm_api_autodeploy.py
🧬 Code graph analysis (1)
tests/integration/defs/accuracy/test_llm_api_autodeploy.py (6)
tensorrt_llm/llmapi/utils.py (1)
get_device_count(135-136)tests/scripts/perf-sanity/run_benchmark_serve.py (1)
llm_models_root(173-174)tests/test_common/llm_data.py (1)
llm_models_root(49-63)tensorrt_llm/llmapi/llm_args.py (4)
world_size(557-558)world_size(567-571)quant_config(2982-2985)quant_config(2988-2989)tensorrt_llm/models/modeling_utils.py (1)
quant_algo(550-551)tensorrt_llm/quantization/mode.py (1)
QuantAlgo(23-48)
⏰ 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)
tests/integration/defs/accuracy/test_llm_api_autodeploy.py (1)
279-298: LGTM - FP8 test implementation follows established patterns.The test correctly:
- Uses device count checking both via marker and runtime check for flexibility with world_size [4, 8]
- Follows the same pattern as
TestNemotronMOE::test_fp8for manually settingquant_configafter LLM initialization- Evaluates both MMLU and GSM8K tasks consistently
tests/integration/test_lists/test-db/l0_dgx_h100.yml (1)
324-325: Test list updates are consistent with code changes, pendingtest_bf16parametrization fix.The test entries correctly reference:
test_bf16[4]- requires the parametrization fix noted earliertest_fp8[4]- correctly matches the parametrized test methodThe 4-GPU condition (lines 309-311) aligns with
world_size=4for both tests.Ensure the
test_bf16parametrization fix is applied; otherwise, test discovery will fail fortest_bf16[4].tests/integration/defs/accuracy/references/mmlu.yaml (1)
351-358: Accuracy threshold adjustments align with PR objectives.The changes lower the MMLU accuracy thresholds for Nemotron-Super-V3:
- BF16: 81.07 → 80.00 (~1.3% decrease)
- FP8 with FP8 KV cache: 78.22 → 77.80 (~0.5% decrease)
This accommodates AutoDeploy testing as stated in the PR description. The NVFP4 entry (lines 356-358) with
kv_cache_quant_algo: FP8remains unchanged and will be used for the FP4 test once enabled.tests/integration/test_lists/test-db/l0_dgx_b200.yml (1)
221-222: Test list updates are consistent with H100 configuration.The AutoDeploy test entries for B200 mirror the H100 configuration:
test_bf16[4]andtest_fp8[4]for TestNemotronSuperV3This ensures consistent test coverage across both GPU platforms.
✏️ Tip: You can disable this entire section by setting review_details to false in your review settings.
|
/bot run |
|
PR_Github #32439 [ run ] triggered by Bot. Commit: |
|
/bot run |
|
PR_Github #32445 [ run ] triggered by Bot. Commit: |
|
PR_Github #32445 [ run ] completed with state
|
galagam
force-pushed
the
gagam/fix-super-accuracy-test
branch
from
6b2b988 to
de37442
Compare
|
/bot run --add-multi-gpu-test |
|
PR_Github #32545 [ run ] triggered by Bot. Commit: |
|
PR_Github #32545 [ run ] completed with state
|
|
/bot run --extra-stage "DGX_B200-4_GPUs-AutoDeploy-1, DGX_H100-4_GPUs-AutoDeploy-1" |
|
PR_Github #32592 [ run ] triggered by Bot. Commit: |
|
PR_Github #32592 [ run ] completed with state
|
- Initial PR NVIDIA#10308 added wrong test name in L0 config files - fix - Add fp8 test - Add (disabled) fp4 test - Slightly decrease bf16 mmlu to accommodate autodeploy test Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com>
Signed-off-by: Gal Hubara Agam <96368689+galagam@users.noreply.github.com>
Signed-off-by: Gal Hubara-Agam <96368689+galagam@users.noreply.github.com>
Signed-off-by: Gal Hubara-Agam <96368689+galagam@users.noreply.github.com>
galagam
force-pushed
the
gagam/fix-super-accuracy-test
branch
from
de37442 to
232200f
Compare
|
/bot run --extra-stage "DGX_B200-4_GPUs-AutoDeploy-1, DGX_H100-4_GPUs-AutoDeploy-1" |
|
PR_Github #32658 [ run ] triggered by Bot. Commit: |
|
PR_Github #32658 [ run ] completed with state
|
|
/bot run --extra-stage "DGX_B200-4_GPUs-AutoDeploy-1, DGX_H100-4_GPUs-AutoDeploy-1" |
|
PR_Github #32709 [ run ] triggered by Bot. Commit: |
|
PR_Github #32709 [ run ] completed with state |
| - accuracy/test_llm_api_autodeploy.py::TestNemotronMOE::test_bf16 | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_bf16 | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_fp8[4] | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_fp8[8] |
There was a problem hiding this comment.
This test config only has 4 GPU. Please remove
There was a problem hiding this comment.
and why are we removing the nano bf16 test?
There was a problem hiding this comment.
This test config only has 4 GPU. Please remove
Right, of course. Will remove.
and why are we removing the nano bf16 test?
It was added by mistake by me, instead of adding the super test.
The nano test is for a single device, so there's no point in adding it here.
@greg-kwasniewski1 is supposed to add a multi-device nano test once it's fixed.
| - accuracy/test_llm_api_autodeploy.py::TestNemotronMOE::test_bf16 | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_bf16 | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_fp8[4] | ||
| - accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_fp8[8] |
4 participants
Description
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.