[TRTLLM-8119][feat] Update doc/tests/chat_template for nano-v2-vlm #8840
Conversation
Wanli-Jiang
requested review from
QiJune,
chzblych,
omera-nv,
rakib-hasan and
symphonylyh
📝 WalkthroughWalkthroughThis pull request introduces support for the NVIDIA-Nemotron-Nano-12B-v2-VL model series. Changes include new documentation, updated model import paths, modifications to prompt token generation, dynamic role assignment in multimodal evaluation, placeholder handling logic for the new model, and expanded test coverage with reference accuracy updates. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~30 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: 3
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tensorrt_llm/inputs/utils.py (1)
568-593: Use Python 3.8-compatible type annotations.Repository guidelines target Python 3.8+. Replace PEP 604 unions and PEP 585 generics in the public signature.
-def apply_chat_template( +def apply_chat_template( *, model_type: str, tokenizer: Union[TransformersTokenizer, TokenizerBase], processor: ProcessorMixin, conversation: list[ConversationMessage], add_generation_prompt: bool, - mm_placeholder_counts: dict[str, int] | list[dict[str, int]], + mm_placeholder_counts: Union[Dict[str, int], List[Dict[str, int]]], tools: Optional[list[dict[str, Any]]] = None, documents: Optional[list[dict[str, str]]] = None, chat_template: Optional[str] = None, chat_template_kwargs: Optional[dict[str, Any]] = None, ) -> (str | List[str]):And, if desired for full 3.8 compliance, switch other local annotations (e.g., list[...] / dict[...]) to List[...] / Dict[...]. As per coding guidelines.
🧹 Nitpick comments (6)
tensorrt_llm/_torch/models/modeling_nemotron_nano.py (1)
519-526: Harden frame-loop alignment and confirm newline intent.
- Use zip(..., strict=True) to fail fast when frame_separators and num_tokens per frame diverge.
- You reintroduce "This is a video:\n". If the model expects no trailing newline after the last frame (per PR intent), please confirm this line doesn't inadvertently add an extra newline to the final chunk.
- for frame_sep, num_tokens in zip(frame_separators, - num_tokens_per_frame): + for frame_sep, num_tokens in zip(frame_separators, + num_tokens_per_frame, strict=True): @@ - for frame_sep in frame_separators: + for frame_sep in frame_separators: # OK as-is + # If you later pair with lengths, prefer: zip(frame_separators, strict=True)Also applies to: 532-539
tensorrt_llm/inputs/utils.py (1)
536-556: Make zip strict to catch misaligned placeholder counts.Add strict=True so content and placeholder counts must match lengths; avoids silent truncation.
- for conv, mm_placeholder_count in zip(conversation, - mm_placeholder_counts): + for conv, mm_placeholder_count in zip(conversation, + mm_placeholder_counts, strict=True):Based on static analysis hints.
examples/models/core/nemotron/README_nano-v2-vl.md (1)
81-84: Tiny grammar nit.Remove the stray space before the period.
- * Prefix-caching is not supported for Nemotron-nano-v2-VL yet . + * Prefix-caching is not supported for Nemotron-nano-v2-VL yet.Optional: if markdownlint is enforced in CI, consider converting bare URLs to text and normalizing list indentation. Based on learnings.
tests/integration/defs/test_e2e.py (1)
2422-2424: Nemotron-Nano V2 VL BF16 coverage looks good; factor model-specific args.Additions and keywords are reasonable. To reduce duplication across tests, consider a small helper that returns extra CLI args for a given model/modality (e.g., max_batch_size, kv-cache flags).
+# Pseudo helper +def _extra_args_for(model_name: str, modality: str) -> list[str]: + args = [] + if model_name == "NVIDIA-Nemotron-Nano-12B-v2-VL-BF16": + args += ["--max_batch_size=128", "--disable_kv_cache_reuse"] + if modality == "video": + args += ["--max_num_tokens=20480"] + return argsThen append cmd.extend(_extra_args_for(model_name, modality)) where used.
Also applies to: 2478-2501, 2598-2605
tests/integration/defs/accuracy/test_llm_api_pytorch.py (2)
3831-3845: Annotate mutable class attributes with ClassVar (fixes RUF012) and tighten max_tokens.
- EXTRA_EVALUATOR_KWARGS, sampling_params, and kv_cache_config are mutable class attributes; annotate with typing.ClassVar.
- Also align MAX_NUM_TOKENS with MMMU.MAX_OUTPUT_LEN to avoid runaway generations on MMMU.
Apply these diffs:
+from typing import ClassVar-class TestNemotron_Nano_12B_V2_VL(LlmapiAccuracyTestHarness): - MODEL_NAME = "nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16" - MODEL_PATH = f"{llm_models_root()}/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16" - MAX_NUM_TOKENS = 25600 - EXTRA_EVALUATOR_KWARGS = dict( - apply_chat_template=True, - system_prompt="/no_think", - ) +class TestNemotron_Nano_12B_V2_VL(LlmapiAccuracyTestHarness): + MODEL_NAME: ClassVar[str] = "nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16" + MODEL_PATH: ClassVar[str] = f"{llm_models_root()}/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16" + MAX_NUM_TOKENS: ClassVar[int] = MMMU.MAX_OUTPUT_LEN + EXTRA_EVALUATOR_KWARGS: ClassVar[dict] = dict( + # apply_chat_template is already True in MMMU.EVALUATOR_KWARGS + system_prompt="/no_think", + )- sampling_params = SamplingParams(max_tokens=MAX_NUM_TOKENS, - truncate_prompt_tokens=MMMU.MAX_INPUT_LEN, - temperature=0.0, - top_k=1, - stop="<|endoftext|>") + sampling_params: ClassVar[SamplingParams] = SamplingParams( + max_tokens=MAX_NUM_TOKENS, + truncate_prompt_tokens=MMMU.MAX_INPUT_LEN, + temperature=0.0, + top_k=1, + stop="<|endoftext|>", + )- kv_cache_config = KvCacheConfig(free_gpu_memory_fraction=0.8, - enable_block_reuse=False) + kv_cache_config: ClassVar[KvCacheConfig] = KvCacheConfig( + free_gpu_memory_fraction=0.8, + enable_block_reuse=False, + )
3831-3834: Minor: drop redundant apply_chat_template override.MMMU.EVALUATOR_KWARGS already sets apply_chat_template=True; overriding it again is redundant. Keep only system_prompt (as above).
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
examples/models/core/nemotron/README_nano-v2-vl.md(1 hunks)examples/models/core/nemotron/README_nemotron-3.md(1 hunks)tensorrt_llm/_torch/models/__init__.py(1 hunks)tensorrt_llm/_torch/models/modeling_nemotron_nano.py(2 hunks)tensorrt_llm/evaluate/lm_eval.py(1 hunks)tensorrt_llm/inputs/utils.py(4 hunks)tests/integration/defs/accuracy/references/mmmu.yaml(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(1 hunks)tests/integration/defs/test_e2e.py(7 hunks)tests/integration/test_lists/qa/llm_function_core.txt(3 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:
tensorrt_llm/evaluate/lm_eval.pytests/integration/defs/test_e2e.pytensorrt_llm/_torch/models/modeling_nemotron_nano.pytests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/_torch/models/__init__.pytensorrt_llm/inputs/utils.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:
tensorrt_llm/evaluate/lm_eval.pytests/integration/defs/test_e2e.pytensorrt_llm/_torch/models/modeling_nemotron_nano.pytests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/_torch/models/__init__.pytensorrt_llm/inputs/utils.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:
tensorrt_llm/evaluate/lm_eval.pytests/integration/defs/test_e2e.pytensorrt_llm/_torch/models/modeling_nemotron_nano.pytests/integration/defs/accuracy/test_llm_api_pytorch.pytensorrt_llm/_torch/models/__init__.pytensorrt_llm/inputs/utils.py
🧠 Learnings (15)
📓 Common learnings
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.
📚 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:
examples/models/core/nemotron/README_nano-v2-vl.mdtests/integration/defs/test_e2e.pyexamples/models/core/nemotron/README_nemotron-3.mdtests/integration/defs/accuracy/test_llm_api_pytorch.pytests/integration/test_lists/qa/llm_function_core.txt
📚 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/defs/test_e2e.pyexamples/models/core/nemotron/README_nemotron-3.mdtests/integration/defs/accuracy/test_llm_api_pytorch.pytests/integration/test_lists/qa/llm_function_core.txt
📚 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/defs/test_e2e.pyexamples/models/core/nemotron/README_nemotron-3.mdtests/integration/test_lists/qa/llm_function_core.txt
📚 Learning: 2025-08-29T14:07:45.863Z
Learnt from: EmmaQiaoCh
Repo: NVIDIA/TensorRT-LLM PR: 7370
File: tests/unittest/trt/model_api/test_model_quantization.py:24-27
Timestamp: 2025-08-29T14:07:45.863Z
Learning: In TensorRT-LLM's CI infrastructure, pytest skip markers (pytest.mark.skip) are properly honored even when test files have __main__ blocks that call test functions directly. The testing system correctly skips tests without requiring modifications to the __main__ block execution pattern.
Applied to files:
tests/integration/defs/test_e2e.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/test_e2e.pyexamples/models/core/nemotron/README_nemotron-3.mdtests/integration/defs/accuracy/test_llm_api_pytorch.pytests/integration/test_lists/qa/llm_function_core.txt
📚 Learning: 2025-08-06T03:47:16.802Z
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6650
File: tests/integration/test_lists/qa/llm_perf_cluster.yml:33-37
Timestamp: 2025-08-06T03:47:16.802Z
Learning: Ministral is a valid model name from Mistral AI, distinct from the regular Mistral models. In TensorRT-LLM test configurations, "ministral_8b" and "ministral_8b_fp8" are correct model identifiers and should not be changed to "mistral_8b".
Applied to files:
tests/integration/defs/test_e2e.py
📚 Learning: 2025-08-18T08:42:02.640Z
Learnt from: samuellees
Repo: NVIDIA/TensorRT-LLM PR: 6974
File: tensorrt_llm/serve/scripts/benchmark_dataset.py:558-566
Timestamp: 2025-08-18T08:42:02.640Z
Learning: In TensorRT-LLM's RandomDataset (tensorrt_llm/serve/scripts/benchmark_dataset.py), when using --random-token-ids option, sequence length accuracy is prioritized over semantic correctness for benchmarking purposes. The encode/decode operations should use skip_special_tokens=True and add_special_tokens=False to ensure exact target token lengths.
Applied to files:
tensorrt_llm/_torch/models/modeling_nemotron_nano.py
📚 Learning: 2025-09-23T15:12:38.312Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/thop/allreduceOp.cpp:352-446
Timestamp: 2025-09-23T15:12:38.312Z
Learning: In TensorRT-LLM NCCL device implementation, NCCL version 2.28+ requirements are handled at runtime in the nccl_device/config layer rather than with compile-time guards. This allows the allreduceOp to remain version-agnostic and delegates version compatibility validation to the appropriate lower-level components that can gracefully handle unsupported configurations.
Applied to files:
examples/models/core/nemotron/README_nemotron-3.md
📚 Learning: 2025-08-01T15:14:45.673Z
Learnt from: yibinl-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 6506
File: examples/models/core/mixtral/requirements.txt:3-3
Timestamp: 2025-08-01T15:14:45.673Z
Learning: In TensorRT-LLM, examples directory can have different dependency versions than the root requirements.txt file. Version conflicts between root and examples dependencies are acceptable because examples are designed to be standalone and self-contained.
Applied to files:
examples/models/core/nemotron/README_nemotron-3.md
📚 Learning: 2025-08-20T07:43:36.447Z
Learnt from: ChristinaZ
Repo: NVIDIA/TensorRT-LLM PR: 7068
File: cpp/tensorrt_llm/kernels/moeTopKFuncs.cuh:169-172
Timestamp: 2025-08-20T07:43:36.447Z
Learning: In TensorRT-LLM MOE kernels, when processing up to 128 experts across 32 threads, each thread handles at most 4 experts (N < 5 constraint), where N represents candidates per thread rather than total system capacity.
Applied to files:
examples/models/core/nemotron/README_nemotron-3.md
📚 Learning: 2025-08-21T00:16:56.457Z
Learnt from: farshadghodsian
Repo: NVIDIA/TensorRT-LLM PR: 7101
File: docs/source/blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.md:36-36
Timestamp: 2025-08-21T00:16:56.457Z
Learning: TensorRT-LLM container release tags in documentation should only reference published NGC container images. The README badge version may be ahead of the actual published container versions.
Applied to files:
examples/models/core/nemotron/README_nemotron-3.md
📚 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:
examples/models/core/nemotron/README_nemotron-3.md
📚 Learning: 2025-10-20T16:54:09.824Z
Learnt from: nvchenghaoz
Repo: NVIDIA/TensorRT-LLM PR: 8469
File: tensorrt_llm/_torch/auto_deploy/custom_ops/rms_norm.py:6-6
Timestamp: 2025-10-20T16:54:09.824Z
Learning: In tensorrt_llm/_torch/auto_deploy/custom_ops/rms_norm.py, the import `from ...modules.mamba.layernorm_gated import _layer_norm_fwd` is correct and should not be changed to modules.fla.layernorm_gated. The _layer_norm_fwd function exists in both modules/mamba/layernorm_gated.py and modules/fla/layernorm_gated.py, but the mamba version is the intended implementation for this use case.
Applied to files:
tensorrt_llm/_torch/models/__init__.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/qa/llm_function_core.txt
🧬 Code graph analysis (3)
tensorrt_llm/evaluate/lm_eval.py (1)
tensorrt_llm/inputs/utils.py (1)
ConversationMessage(427-435)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (4)
tests/integration/defs/accuracy/accuracy_core.py (3)
MMMU(386-403)evaluate(184-247)evaluate(766-776)tests/integration/defs/conftest.py (1)
llm_models_root(80-94)tensorrt_llm/evaluate/lm_eval.py (4)
apply_chat_template(66-78)apply_chat_template(197-249)MMMU(662-715)evaluate(394-429)tensorrt_llm/llmapi/llm_args.py (1)
KvCacheConfig(1265-1409)
tensorrt_llm/_torch/models/__init__.py (1)
tensorrt_llm/_torch/models/modeling_nemotron_nano.py (1)
NemotronH_Nano_VL_V2(682-855)
🪛 markdownlint-cli2 (0.18.1)
examples/models/core/nemotron/README_nano-v2-vl.md
4-4: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
4-4: Bare URL used
(MD034, no-bare-urls)
5-5: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
5-5: Bare URL used
(MD034, no-bare-urls)
6-6: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
6-6: Bare URL used
(MD034, no-bare-urls)
9-9: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
10-10: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
11-11: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
12-12: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
13-13: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
14-14: Unordered list indentation
Expected: 0; Actual: 2
(MD007, ul-indent)
18-18: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
20-20: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
26-26: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
32-32: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
38-38: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
46-46: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
82-82: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
83-83: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
84-84: Unordered list indentation
Expected: 0; Actual: 1
(MD007, ul-indent)
🪛 Ruff (0.14.2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py
3831-3834: Mutable class attributes should be annotated with typing.ClassVar
(RUF012)
tensorrt_llm/inputs/utils.py
537-538: 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 (8)
tensorrt_llm/inputs/utils.py (1)
378-378: LGTM: model added to placeholder exceptions.Including "NemotronH_Nano_VL_V2" here aligns with the new conversion path and custom content assembly.
examples/models/core/nemotron/README_nemotron-3.md (1)
1-1: LGTM.Title update is consistent with file scope.
tests/integration/defs/accuracy/references/mmmu.yaml (1)
3-7: Record provenance for the new MMMU number.Please add the eval date, commit/tag of the model, and harness version flags used so we can reproduce 26.67 later.
tests/integration/defs/test_e2e.py (1)
2841-2845: FP8 path: skip and flags look appropriate.
- Skipping video for FP8 chunked prefill is explicitly handled; good.
- Disabling KV cache reuse and bounding max_batch_size are consistent with BF16.
If FP8 enables chunked prefill video later, please drop the skip and tune max_num_tokens accordingly.
Also applies to: 2863-2869, 3005-3010
tensorrt_llm/_torch/models/__init__.py (1)
20-20: No stale imports detected—migration is clean.The grep search confirms that the old module path
modeling_nanov2vlmhas been completely removed. All remaining references toNemotronH_Nano_VL_V2are consistent: string-based model type checks in utils.py, the correct new import in__init__.py, class definition inmodeling_nemotron_nano.py, and documentation. No split sources or stale imports remain.tests/integration/test_lists/qa/llm_function_core.txt (3)
612-612: Symbol rename alignment looks good.Updated to TestNemotron_Nano_12B_V2_VL::test_auto_dtype; matches the new class.
656-661: New Nemotron Nano 12B V2 VL multimodal e2e entries — OK.The new image/video/mixture_text_image cases look consistent with existing patterns.
690-690: FP8 chunked prefill entry for Nemotron Nano 12B V2 VL — OK.Name/path format matches the existing suite.
|
/bot run |
|
PR_Github #23177 [ run ] triggered by Bot. Commit: |
|
PR_Github #23177 [ run ] completed with state |
Wanli-Jiang
force-pushed
the
user/williamj/update-nanov2vlm
branch
3 times, most recently
from
102c720 to
cee6a3c
Compare
|
/bot run |
|
PR_Github #23643 [ run ] triggered by Bot. Commit: |
|
PR_Github #23643 [ run ] completed with state |
|
/bot run |
|
PR_Github #23694 [ run ] triggered by Bot. Commit: |
|
PR_Github #23694 [ run ] completed with state |
Wanli-Jiang
force-pushed
the
user/williamj/update-nanov2vlm
branch
2 times, most recently
from
c199033 to
1233001
Compare
|
/bot run |
|
PR_Github #23800 [ run ] triggered by Bot. Commit: |
|
PR_Github #23800 [ run ] completed with state |
yechank-nvidia
left a comment
yechank-nvidia
left a comment
There was a problem hiding this comment.
Thx for adding the unittest. Left comment about deletion of e2e test.
Wanli-Jiang
force-pushed
the
user/williamj/update-nanov2vlm
branch
from
1233001 to
2ded563
Compare
Signed-off-by: Wanli Jiang <35160485+Wanli-Jiang@users.noreply.github.com>
Wanli-Jiang
force-pushed
the
user/williamj/update-nanov2vlm
branch
from
2ded563 to
5fe0107
Compare
|
/bot run |
|
PR_Github #24022 [ run ] triggered by Bot. Commit: |
|
PR_Github #24022 [ run ] completed with state |
|
/bot run |
|
PR_Github #24088 [ run ] triggered by Bot. Commit: |
|
PR_Github #24088 [ run ] completed with state |
|
/bot run |
|
PR_Github #24139 [ run ] triggered by Bot. Commit: |
|
PR_Github #24139 [ run ] completed with state |
|
/bot run |
|
PR_Github #24188 [ run ] triggered by Bot. Commit: |
|
PR_Github #24188 [ run ] completed with state |
…VIDIA#8840) Signed-off-by: Wanli Jiang <35160485+Wanli-Jiang@users.noreply.github.com>
3 participants
Feature:
Summary by CodeRabbit
New Features
Bug Fixes
Tests
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.