[None][doc] Add Qwen3 Next Quick Start Guide #8195
Conversation
|
/bot run |
Funatiq
changed the title
Funatiq
force-pushed
the
dev/qwen3-next-guide
branch
from
14293a4 to
f34375f
Compare
|
PR_Github #20786 [ run ] triggered by Bot |
Signed-off-by: Faradawn Yang <[email protected]>
Signed-off-by: Faradawn Yang <[email protected]>
- Updated L0_Test.groovy to include auto_trigger for Qwen3Next. - Added TestQwen3NextThinking class in test_llm_api_pytorch.py for accuracy testing. - Updated l0_dgx_h100.yml to include new test case for Qwen3Next under specific conditions. Signed-off-by: Robin Kobus <[email protected]>
Funatiq
force-pushed
the
dev/qwen3-next-guide
branch
from
f34375f to
5596ca7
Compare
|
/bot run |
|
PR_Github #20787 [ run ] triggered by Bot |
|
PR_Github #20786 [ run ] completed with state |
|
PR_Github #20787 [ run ] completed with state |
|
/bot run |
|
PR_Github #20870 [ run ] triggered by Bot |
|
PR_Github #20870 [ run ] completed with state |
|
/bot skip --comment "Unrelated test failure" |
📝 WalkthroughWalkthroughAdds Qwen3-Next documentation (quick-start for TensorRT-LLM and supported models entries), extends CI to recognize and launch Qwen3Next-specific Jenkins stages, and introduces accuracy references and an integration test for Qwen3-Next-80B-A3B-Thinking with corresponding test gating on DGX H100 (4 GPUs). Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Dev as Developer
participant Git as Repo
participant J as Jenkins
participant M as getMakoArgsFromStageName
participant L as launchTestJobs
Dev->>Git: Push/PR with stage name containing "-Qwen3Next-"
J->>M: Resolve MakO args for stage name
M-->>J: auto_trigger = qwen3_next
J->>L: Launch tests with auto_trigger=qwen3_next
L-->>J: Queue DGX_H100-4_GPUs-PyTorch-Qwen3Next-1
J->>J: Run TestQwen3NextThinking (tp4, pp1, ep4)
J-->>Dev: Report test results
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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 (3)
docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md (2)
24-30: Add language specifier to code block.The code block starting at line 24 should specify
shellas the language for proper syntax highlighting and rendering.Apply this diff:
-``` +```shell cd TensorRT-LLM
176-178: Add language specifier to code block.The response example should specify
jsonas the language for proper syntax highlighting.Apply this diff:
-``` +```json {"id":"chatcmpl-64ac201c77bf46a7a3a4eca7759b1fd8",...}tests/integration/test_lists/test-db/l0_dgx_h100.yml (1)
171-185: Consider adding orchestrator term for consistency.The new Qwen3Next gating block is missing the
orchestrator: mpiterm. While the Jenkins pipeline will add this by default (line 1291 injenkins/L0_Test.groovy), explicitly including it would improve consistency with other similar blocks at lines 14-16 (auto_trigger: others), 103-104 (auto_trigger: deepseek), and 161-162 (auto_trigger: gpt_oss).Apply this diff:
terms: stage: pre_merge backend: pytorch auto_trigger: qwen3_next + orchestrator: mpi tests:
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (7)
docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md(1 hunks)docs/source/models/supported-models.md(2 hunks)jenkins/L0_Test.groovy(2 hunks)tests/integration/defs/accuracy/references/gsm8k.yaml(1 hunks)tests/integration/defs/accuracy/references/mmlu.yaml(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(1 hunks)tests/integration/test_lists/test-db/l0_dgx_h100.yml(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/integration/defs/accuracy/test_llm_api_pytorch.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/integration/defs/accuracy/test_llm_api_pytorch.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/integration/defs/accuracy/test_llm_api_pytorch.py
🧠 Learnings (1)
📚 Learning: 2025-08-26T09:49:04.956Z
Learnt from: pengbowang-nv
PR: NVIDIA/TensorRT-LLM#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.yml
🧬 Code graph analysis (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (4)
tests/integration/defs/accuracy/accuracy_core.py (5)
LlmapiAccuracyTestHarness(844-855)MMLU(315-329)evaluate(184-245)evaluate(763-773)GSM8K(332-347)tests/integration/defs/conftest.py (2)
llm_models_root(78-92)get_device_count(1978-1980)tensorrt_llm/llmapi/llm_args.py (2)
KvCacheConfig(1106-1240)CudaGraphConfig(109-166)tensorrt_llm/llmapi/llm.py (1)
LLM(1084-1100)
🪛 markdownlint-cli2 (0.18.1)
docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md
24-24: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
176-176: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🔇 Additional comments (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
3563-3592: Verify the max_batch_size configuration.The test implementation follows established patterns correctly and properly integrates with the test harness. However,
max_batch_size=720(line 3579) is significantly higher than similar models:
- Qwen3-30B-A3B uses
max_batch_size=32(line 2883)- Qwen3-8B uses
max_batch_size=64(line 2673)For an 80B model with 60% KV cache memory allocation, verify whether:
- 720 is intentional for specific throughput testing scenarios
- This should be a lower value to avoid potential OOM issues in CI
- This is a typo (e.g., should be 72)
Otherwise, the implementation is clean and consistent with the codebase patterns.
| --result-filename "concurrency_${concurrency}.json" | ||
| ``` | ||
|
|
||
| For more benchmarking options see <https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt\_llm/serve/scripts/benchmark\_serving.py>. |
There was a problem hiding this comment.
Fix broken URL with escaped underscores.
The URL contains escaped underscores (\_) which will break the link. These should be regular underscores.
Apply this diff:
-For more benchmarking options see <https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt\_llm/serve/scripts/benchmark\_serving.py>.
+For more benchmarking options see <https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/serve/scripts/benchmark_serving.py>.📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| For more benchmarking options see <https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt\_llm/serve/scripts/benchmark\_serving.py>. | |
| For more benchmarking options see <https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/serve/scripts/benchmark_serving.py>. |
🤖 Prompt for AI Agents
In docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md
around line 237 the link contains escaped underscores (`\_`) which break the
URL; replace each `\_` with a plain underscore so the URL reads
https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/serve/scripts/benchmark_serving.py
(update the markdown to use the corrected, unescaped underscores).
There was a problem hiding this comment.
Seems like the humans are having a chat. I'll hop back into my burrow for now. If you need me again, just tag @coderabbitai in a new comment, and I'll come hopping out!
|
PR_Github #20889 [ skip ] triggered by Bot |
|
PR_Github #20889 [ skip ] completed with state |
docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md
Show resolved
Hide resolved
docs/source/deployment-guide/quick-start-recipe-for-qwen3-next-on-trtllm.md
Outdated
Show resolved
Hide resolved
|
/bot run --stage-list "DGX_H100-4_GPUs-PyTorch-Others-1" |
|
PR_Github #21021 [ run ] triggered by Bot |
|
PR_Github #21021 [ run ] completed with state |
- Remove Qwen3Next stage. - Add Qwen3Next test to PyTorch-Others stage instead. Signed-off-by: Robin Kobus <[email protected]>
Signed-off-by: Robin Kobus <[email protected]>
Funatiq
force-pushed
the
dev/qwen3-next-guide
branch
from
9aed666 to
10f48af
Compare
|
/bot run --stage-list "DGX_H100-4_GPUs-PyTorch-Others-1" |
|
PR_Github #21031 [ run ] triggered by Bot |
|
PR_Github #21031 [ run ] completed with state |
|
|
||
| ## Introduction | ||
|
|
||
| This is a functional quick-start guide for running the Qwen3-Next model on TensorRT LLM. It focuses on a working setup with recommended defaults. Additional performance optimizations and support (such as Blackwell) will be rolled out in future updates. |
There was a problem hiding this comment.
Let's remove such on Blackwell
|
|
||
| ## Prerequisites | ||
|
|
||
| * GPU: NVIDIA Hopper Architecture |
There was a problem hiding this comment.
GPU: NVIDIA Blackwell and Hopper Architecture
| trtllm-serve Qwen/Qwen3-Next-80B-A3B-Thinking \ | ||
| --host 0.0.0.0 \ | ||
| --port 8000 \ | ||
| --backend pytorch \ |
There was a problem hiding this comment.
No need to specify --bakcend Pytorch explicitly.
| * **Description:** A value between `0.0` and `1.0` that specifies the fraction of free GPU memory to reserve for the KV cache after the model is loaded. Since memory usage can fluctuate, this buffer helps prevent out-of-memory (OOM) errors. | ||
| * **Recommendation:** If you experience OOM errors, try reducing this value to `0.7` or lower. | ||
|
|
||
| #### `--backend pytorch` |
| - accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp2pp2-fp8kv=True-attn_backend=TRTLLM-torch_compile=False] | ||
| - accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp2pp2-fp8kv=True-attn_backend=TRTLLM-torch_compile=True] | ||
| - accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_guided_decoding_4gpus[xgrammar] | ||
| - accuracy/test_llm_api_pytorch.py::TestQwen3NextThinking::test_auto_dtype[tp4ep4] |
There was a problem hiding this comment.
I prefer to put this case under l0_dgx_b200.yml
|
I saw that the new cherry pick PR includes all the changes: #8288. Do we need to modify this PR? |
6 participants
Summary by CodeRabbit
Documentation
Tests
Chores
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.
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.