[TRTLLM-6657][feat] Add LoRA support for Gemma3 #6371
Conversation
📝 WalkthroughWalkthroughThe changes replace a custom MLP implementation in the Gemma3 model with a standardized gated MLP module using a specific GELU activation variant. LoRA parameter support is added throughout model forward passes. Activation handling in the GatedMLP module is generalized. Integration and performance test configurations are updated to support a new model variant and a dummy LoRA test. Changes
Sequence Diagram(s)LoRA Parameter Forwarding in Gemma3 ModelsequenceDiagram
participant User
participant Gemma3TextModel
participant Gemma3DecoderLayer
participant Attention
participant GatedMLP
User->>Gemma3TextModel: forward(..., lora_params)
Gemma3TextModel->>Gemma3DecoderLayer: forward(..., lora_params)
Gemma3DecoderLayer->>Attention: forward(..., lora_params)
Gemma3DecoderLayer->>GatedMLP: forward(..., lora_params)
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~15–20 minutes Possibly related PRs
Suggested reviewers
📜 Recent review detailsConfiguration used: .coderabbit.yaml 📒 Files selected for processing (7)
✅ Files skipped from review due to trivial changes (3)
🚧 Files skipped from review as they are similar to previous changes (4)
⏰ 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)
✨ 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. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
1df218f to
8151fd0
Compare
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
3 times, most recently
from
5cfd6c9 to
2bbe9b6
Compare
|
/bot run |
|
PR_Github #13034 [ run ] triggered by Bot |
![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
🧹 Nitpick comments (2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (2)
618-618: Fix line length violation.The comment line exceeds the 120-character limit specified in the coding guidelines.
Break the comment into multiple lines to comply with the line length limit:
- # Disabling kv cache reuse as a WAR to deal with gaps in kernel support for Gemma3's non-inclusive sliding window size. + # Disabling kv cache reuse as a WAR to deal with gaps in kernel support + # for Gemma3's non-inclusive sliding window size.
606-607: Consider documenting future MLP module support.The commented-out MLP modules (
gate_proj,down_proj,up_projand their TensorRT-LLM equivalentsmlp_h_to_4h,mlp_4h_to_h,mlp_gate) suggest planned expansion of LoRA support beyond attention modules.Consider adding a TODO comment to document the planned expansion:
target_modules=["q_proj", "k_proj", "v_proj"], # "gate_proj", "down_proj", "up_proj"], + # TODO: Add MLP modules when LoRA support for GatedMLP is fully implementedAlso applies to: 614-615
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py(2 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, 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. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (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 interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved 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.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
🪛 Ruff (0.12.2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py
618-618: Line too long (127 > 120)
(E501)
⏰ 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 (6)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (6)
26-27: LGTM! Import additions are appropriate.The new imports for LoRA functionality are correctly placed and both are utilized in the new test method.
595-596: Good documentation and method naming.The comment clearly indicates this is a smoke test for LoRA functionality, and the method name follows the established naming convention.
597-608: Well-structured LoRA setup with appropriate test parameters.The dummy LoRA generation uses reasonable parameters (rank 32, zero weights for smoke testing) and targets attention modules as a good starting point for LoRA integration testing.
610-617: LoRA configuration looks correct with proper module mapping.The mapping from HuggingFace module names (
q_proj,k_proj,v_proj) to TensorRT-LLM module names (attn_q,attn_k,attn_v) appears to be correct based on the model architecture.
619-622: Appropriate workaround configuration documented.The KV cache configuration with disabled block and partial reuse is properly documented as a workaround for kernel support gaps.
623-625: Test execution follows established patterns.The LLM instantiation with LoRA configuration and GSM8K task evaluation follows the established testing pattern in this file.
brb-nv
requested review from
shaharmor98
and removed request for
dc3671,
litaotju,
byshiue and
liji-nv
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
cb175f1 to
b3a4fde
Compare
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
623-625: LGTM: LLM instantiation and evaluation are correct.The test properly instantiates the LLM with LoRA configuration and runs a meaningful evaluation task to verify functionality.
Note: This addresses the past review comment about LoRA testing approach. The test creates zero-weighted LoRAs as a smoke test, which is appropriate for verifying that the LoRA infrastructure works without affecting model outputs. The GSM8K evaluation ensures the model runs successfully with LoRA enabled.
🧹 Nitpick comments (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
618-618: Fix line length violation.The line exceeds the 120-character limit. Consider breaking it into multiple lines for better readability.
- # Disabling kv cache reuse as a WAR to deal with gaps in kernel support for Gemma3's non-inclusive sliding window size. + # Disabling kv cache reuse as a WAR to deal with gaps in kernel support + # for Gemma3's non-inclusive sliding window size.
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tensorrt_llm/_torch/models/modeling_gemma3.py(7 hunks)tensorrt_llm/_torch/models/modeling_gemma3vl.py(1 hunks)tensorrt_llm/_torch/modules/gated_mlp.py(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(2 hunks)
🚧 Files skipped from review as they are similar to previous changes (3)
- tensorrt_llm/_torch/models/modeling_gemma3vl.py
- tensorrt_llm/_torch/modules/gated_mlp.py
- tensorrt_llm/_torch/models/modeling_gemma3.py
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, 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. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (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 interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved 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.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
🧠 Learnings (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
Learnt from: amitz-nv
PR: #5616
File: tensorrt_llm/executor/worker.py:375-384
Timestamp: 2025-07-17T09:01:27.402Z
Learning: In tensorrt_llm/executor/worker.py, the LoRA adapter cache optimization logic that checks is_adapter_in_cpu_cache() and conditionally passes None for weights/config has a known race condition issue that cannot be solved with simple error handling or verification checks. This is a known limitation that requires a more comprehensive solution.
🪛 Ruff (0.12.2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py
618-618: Line too long (127 > 120)
(E501)
🔇 Additional comments (5)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (5)
26-27: LGTM: Import statements are appropriate.The new imports for
generate_dummy_lorasandLoraConfigare necessary for the LoRA testing functionality and follow the existing import conventions.
595-596: LGTM: Clear test documentation.The comment clearly identifies this as a smoke test for LoRA functionality, which sets appropriate expectations.
597-608: LGTM: LoRA generation parameters are appropriate.The test uses reasonable parameters:
- Rank 32 is a common LoRA rank for testing
- Single LoRA keeps the test simple
- Zero weights ensure safety during testing
- Target modules cover the key attention projections
619-622: LGTM: KV cache configuration is appropriate.The KV cache configuration correctly disables block and partial reuse as a workaround for Gemma3's sliding window limitations, consistent with other Gemma3 tests in the file.
610-617: Mapping consistency confirmed
The default mapping intensorrt_llm/lora_manager.py:get_default_trtllm_modules_to_hf_modules()shows:
- “attn_q” → “q_proj”
- “attn_k” → “k_proj”
- “attn_v” → “v_proj”
This matches the HF modules
["q_proj","k_proj","v_proj"]and TRTLLM modules["attn_q","attn_k","attn_v"]used in the test. No changes required.
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
b3a4fde to
e618782
Compare
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (2)
621-621: Fix line length violation.The line exceeds the 120-character limit specified in the static analysis hints.
- # Disabling kv cache reuse as a WAR to deal with gaps in kernel support for Gemma3's non-inclusive sliding window size. + # Disabling kv cache reuse as a WAR to deal with gaps in kernel support for + # Gemma3's non-inclusive sliding window size.
595-632: Consider expanding test coverage for LoRA functionality.While this smoke test is valuable for verifying basic LoRA integration, consider adding tests that:
- Verify LoRA weights are actually applied (non-zero weight LoRAs)
- Test multiple LoRA adapters if supported
- Validate LoRA ID specification mechanisms mentioned in past comments
This would provide more comprehensive coverage of the LoRA feature.
Would you like me to help design additional LoRA test cases that address the questions raised in the past review comments about LoRA ID specification and CPU LoRA handling?
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
tensorrt_llm/_torch/models/modeling_gemma3.py(7 hunks)tensorrt_llm/_torch/models/modeling_gemma3vl.py(1 hunks)tensorrt_llm/_torch/modules/gated_mlp.py(1 hunks)tests/integration/defs/accuracy/test_llm_api_pytorch.py(3 hunks)
🚧 Files skipped from review as they are similar to previous changes (3)
- tensorrt_llm/_torch/models/modeling_gemma3vl.py
- tensorrt_llm/_torch/modules/gated_mlp.py
- tensorrt_llm/_torch/models/modeling_gemma3.py
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: Python code should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, 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. Prefix k for variable names that start with a number (e.g., k_99th_percentile).
Python global variables should use upper snake_case and prefix G (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 interfaces that may be used outside a Python file, prefer docstrings over comments.
Comments in Python should be reserved 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.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
tests/integration/defs/accuracy/test_llm_api_pytorch.py
🧠 Learnings (1)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (1)
Learnt from: amitz-nv
PR: #5616
File: tensorrt_llm/executor/worker.py:375-384
Timestamp: 2025-07-17T09:01:27.402Z
Learning: In tensorrt_llm/executor/worker.py, the LoRA adapter cache optimization logic that checks is_adapter_in_cpu_cache() and conditionally passes None for weights/config has a known race condition issue that cannot be solved with simple error handling or verification checks. This is a known limitation that requires a more comprehensive solution.
🪛 Ruff (0.12.2)
tests/integration/defs/accuracy/test_llm_api_pytorch.py
621-621: Line too long (127 > 120)
(E501)
⏰ 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 (6)
tests/integration/defs/accuracy/test_llm_api_pytorch.py (6)
18-18: LGTM! Import statements are correctly added.The imports for
generate_dummy_lorasandLoraConfigare appropriate for the LoRA functionality being tested and follow Python naming conventions.Also applies to: 27-27
595-596: Good smoke test implementation.The comment clearly indicates this is a smoke test to verify LoRA functionality works without errors, which is appropriate for integration testing.
597-610: LoRA generation configuration looks correct.The dummy LoRA generation with zero weights targeting
q_proj,k_proj, andv_projmodules is appropriate for testing. The parameters (rank 32, zero weights) ensure the test focuses on integration rather than functional correctness.
611-620: LoRA configuration mapping is correct.The mapping between HuggingFace module names (
q_proj,k_proj,v_proj) and TensorRT-LLM target modules (attn_q,attn_k,attn_v) is appropriate and consistent with the model architecture.
622-625: KV cache configuration is consistent.The KV cache configuration matches the pattern used in other Gemma3 tests in this class, appropriately disabling block and partial reuse as a workaround for sliding window size issues.
626-632: LLM instantiation and evaluation are appropriate.The LLM is correctly configured with LoRA settings and the GSM8K evaluation provides a meaningful test of the LoRA-enabled model functionality.
|
PR_Github #13034 [ run ] completed with state |
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
2 times, most recently
from
971a247 to
5d962e8
Compare
Yup. Ran it locally. Adding it here: https://gitlab-master.nvidia.com/ftp/llm-models/-/merge_requests/370 |
|
/bot run |
|
PR_Github #13730 [ run ] triggered by Bot |
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
5d962e8 to
0ec13a3
Compare
|
/bot run |
|
PR_Github #13738 [ run ] triggered by Bot |
|
PR_Github #13730 [ run ] completed with state |
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
0ec13a3 to
38043e8
Compare
Signed-off-by: Balaram Buddharaju <[email protected]>
brb-nv
force-pushed
the
user/brb/lora-for-gemma3
branch
from
38043e8 to
b54d7df
Compare
brb-nv
changed the title
|
/bot run |
|
PR_Github #13738 [ run ] completed with state |
|
PR_Github #13776 [ run ] triggered by Bot |
|
@brb-nv can you add an unittest for lora? Example: https://github.com/NVIDIA/TensorRT-LLM/blob/main/tests/unittest/llmapi/test_llm_pytorch.py#L446 |
|
PR_Github #13776 [ run ] completed with state |
Hi Wanli, sorry I set the MR to auto-merge so that it merged overnight (CI is taking very long making room for frequent merge conflicts). I'll create a follow-up MR adding the unit test. |
Signed-off-by: Balaram Buddharaju <[email protected]> Signed-off-by: Lanyu Liao <[email protected]>
Signed-off-by: Balaram Buddharaju <[email protected]>
Description
This MR adds text-only LoRA support for Gemma3.
Previously, I was using a custom MLP implementation to accommodate the activation function
pytorch_gelu_tanh. Now, I'm switching to TRTLLM's GatedMLP so that we can take advantage of LoRA integration there.Test Coverage
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.
Summary by CodeRabbit
New Features
Bug Fixes
Tests