[None][feat] add model seed-oss #7496
Conversation
📝 WalkthroughWalkthroughAdds a new Torch-based SeedOss causal language model implementation and exposes it publicly. Introduces attention, decoder layer, model, and causal LM wrapper classes, integrates positional embedding configuration, input validation for embeddings vs. IDs, and metadata handling. Updates package init to import and export SeedOssForCausalLM. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant U as User
participant LM as SeedOssForCausalLM
participant M as SeedOssModel
participant L as SeedOssDecoderLayer*
participant A as SeedOssAttention
participant P as GatedMLP
U->>LM: generate()/forward(input_ids or inputs_embeds,<br/>attn_metadata, mrope_config?, spec_metadata?)
LM->>M: forward(...)
alt XOR input selection
M->>M: compute inputs_embeds from input_ids
else
M->>M: use provided inputs_embeds
end
loop for each layer
M->>L: forward(hidden_states, residual,<br/>position_ids, attn_metadata,<br/>mrope_config?, spec_metadata?)
L->>A: self-attn(hidden_states, attn_metadata,<br/>rope params)
A-->>L: attn_output
L->>P: mlp(post-attn normalized states)
P-->>L: mlp_output
note over L: maybe_capture_hidden_states(...) if spec_metadata
L-->>M: (hidden_states, residual)
end
M->>M: final RMSNorm
M-->>LM: hidden_states
LM-->>U: logits / outputs
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes ✨ 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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tensorrt_llm/_torch/models/__init__.py (1)
1-1: Add NVIDIA copyright header.Per repo guidelines, prepend the current-year NVIDIA copyright header to all Python sources.
Add at the very top:
# Copyright (c) 2025, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
🧹 Nitpick comments (3)
tensorrt_llm/_torch/models/modeling_seedoss.py (3)
56-56: Fix init return annotation (must be None).
__init__should not annotate a tuple return type.Apply this diff:
- ) -> Tuple[torch.Tensor, torch.Tensor]: + ) -> None:
79-88: Fix forward() return annotation to match actual return value.Method returns
(hidden_states, residual)but annotation saystorch.Tensor.Apply this diff:
- ) -> torch.Tensor: + ) -> Tuple[torch.Tensor, torch.Tensor]:
153-156: Clarify the input validation and message.Logic is fine, but the message doesn’t cover the “both None” case. Slightly clearer predicate + message.
Apply this diff:
- if (input_ids is None) ^ (inputs_embeds is not None): - raise ValueError( - "You cannot specify both input_ids and inputs_embeds at the same time, and must specify either one" - ) + if (input_ids is None) == (inputs_embeds is None): + raise ValueError("Specify exactly one of input_ids or inputs_embeds.")
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (2)
tensorrt_llm/_torch/models/__init__.py(2 hunks)tensorrt_llm/_torch/models/modeling_seedoss.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (4)
**/*
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Filenames compiled into a target must be case-insensitively unique
Files:
tensorrt_llm/_torch/models/__init__.pytensorrt_llm/_torch/models/modeling_seedoss.py
**/*.{h,hpp,hh,hxx,cc,cpp,cxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use spaces, not tabs; indent 4 spaces
Files:
tensorrt_llm/_torch/models/__init__.pytensorrt_llm/_torch/models/modeling_seedoss.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Code must target Python 3.8+
Indent with 4 spaces; do not use tabs (Python)
Maintain module namespace on import: prefer from package.subpackage import foo; use foo.Symbol()
Python filenames use snake_case
Python class names use PascalCase
Python functions and methods use snake_case
Python local variables use snake_case; if starting with a number concept, prefix with k (e.g., k_99th_percentile)
Python global variables use G_ prefix with UPPER_SNAKE_CASE
Python constants use UPPER_SNAKE_CASE
Avoid shadowing variables from outer scopes
Initialize all externally visible class members in init
For public interfaces, prefer docstrings over comments; comments should be for in-function or file-local interfaces
Use Google-style docstrings for classes and functions (Sphinx-parsable)
Document attributes and variables inline with docstrings immediately after assignment
Avoid reflection when a non-reflective approach suffices
Limit except clauses to specific exceptions where possible
When using try/except for duck-typing, keep try body minimal and move logic to else
Files:
tensorrt_llm/_torch/models/__init__.pytensorrt_llm/_torch/models/modeling_seedoss.py
**/*.{cpp,cc,cxx,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend NVIDIA copyright header (current year) to all source files
Files:
tensorrt_llm/_torch/models/__init__.pytensorrt_llm/_torch/models/modeling_seedoss.py
🧬 Code graph analysis (2)
tensorrt_llm/_torch/models/__init__.py (1)
tensorrt_llm/_torch/models/modeling_seedoss.py (1)
SeedOssForCausalLM(179-188)
tensorrt_llm/_torch/models/modeling_seedoss.py (7)
tensorrt_llm/_torch/attention_backend/interface.py (3)
AttentionMetadata(39-331)PositionalEmbeddingParams(501-516)RopeParams(345-497)tensorrt_llm/_torch/modules/embedding.py (1)
Embedding(164-242)tensorrt_llm/_torch/modules/gated_mlp.py (1)
GatedMLP(18-164)tensorrt_llm/_torch/modules/linear.py (1)
TensorParallelMode(44-56)tensorrt_llm/_torch/speculative/interface.py (1)
SpecMetadata(116-211)tensorrt_llm/_torch/models/modeling_speculative.py (1)
SpecDecOneEngineForCausalLM(360-468)tensorrt_llm/_torch/models/modeling_utils.py (1)
register_auto_model(595-601)
svc-trtllm-gh-bot
added
the
Community want to contribute
|
@Nekofish-L , thank you for the contribution. |
Nekofish-L
force-pushed
the
dev-seed-oss
branch
3 times, most recently
from
6e10d17 to
cd3ef55
Compare
|
Hi @karljang , Thank you for the review! |
|
@Nekofish-L , |
Nekofish-L
force-pushed
the
dev-seed-oss
branch
from
cd3ef55 to
2092a12
Compare
|
Hi @karljang , |
Nekofish-L
force-pushed
the
dev-seed-oss
branch
from
21fd559 to
3d7b740
Compare
|
Batch Size Under Different token/s SLOs (H20-96G)
acc:
|
|
@Nekofish-L , there is a coding style check failed. Would you be able to address that here: https://github.com/NVIDIA/TensorRT-LLM/actions/runs/17551165567/job/49982861443?pr=7496 |
|
Hi @karljang , |
|
@Wanli-Jiang ，this PR requires to upgrade transformers to 4.56. Could you please upgrade transformers in a separate PR ? |
#7523, some CI tests failed, need to fix them firstly. Let's wait for a while. |
|
Hi @Wanli-Jiang , could you please review this PR? Its dependency(#7523 ) has now been merged into the main branch. |
Wanli-Jiang
left a comment
Wanli-Jiang
left a comment
There was a problem hiding this comment.
codes are neat and look good to me.
@nv-guomingz should we require the unittests for community contributed models?
|
@Nekofish-L can you rebase your commits on recent main branch? It is conflicted with ToT. thanks! about the tests, other teammates will help. |
Signed-off-by: Nekofish-L <[email protected]>
Signed-off-by: Nekofish-L <[email protected]>
Signed-off-by: Nekofish-L <[email protected]>
Signed-off-by: Nekofish-L <[email protected]>
Signed-off-by: Nekofish-L <[email protected]>
Nekofish-L
force-pushed
the
dev-seed-oss
branch
from
43e51bd to
84ed53d
Compare
|
@Wanli-Jiang , I've rebased my branch onto the latest main and resolved the conflicts. It should be good to go now. |
|
/bot run |
Wanli-Jiang
left a comment
Wanli-Jiang
left a comment
There was a problem hiding this comment.
thanks for the contribution
|
PR_Github #19772 [ run ] triggered by Bot |
|
PR_Github #19772 [ run ] completed with state |
6 participants
Summary by CodeRabbit
Description
This PR adds support for the seed-oss model inference in TensorRT-LLM pytorch backend.
Dependency
transformers >= 4.56.0Related Issues
#7196
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.