[#8733][feat] Add Llama4 MoE handling to AutoDeploy #9556
Merged
+1,354
−100
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Tal Cherckez <[email protected]>
tcherckez-nvidia
changed the title
tcherckez-nvidia
changed the title
Contributor
📝 WalkthroughWalkthroughThis PR introduces support for BMM-based Mixture of Experts (MoE) pattern matching and fusion with stacked TRT-LLM formatted weights. It adds conditional routing modes (input-side vs. output-side), new custom operators, pattern detection for Llama4-style MoE, and corresponding tensor parallelism sharding support. Changes
Sequence DiagramsequenceDiagram
participant Graph as Computation Graph
participant PatternMatcher as MatchBmmMoePattern
participant TransformLib as Transform Library
participant CustomOp as Custom Operators
participant Sharding as Sharding Engine
Graph->>PatternMatcher: Scan for BMM-based MoE pattern
PatternMatcher->>PatternMatcher: Locate gate+up BMM operations
PatternMatcher->>PatternMatcher: Extract input/routing and output/routing
PatternMatcher->>TransformLib: Normalize weights (Llama4 → TRT-LLM format)
TransformLib->>TransformLib: Transpose w3_w1_stacked and w2_stacked
TransformLib->>CustomOp: Insert torch_moe_bmm call with stacked weights
CustomOp->>CustomOp: Apply input-side routing (scale input by routing_weights)
CustomOp->>CustomOp: Invoke per-expert MLPs
CustomOp->>Graph: Return fused MoE output
Graph->>Sharding: Detect torch_moe_bmm for EP sharding
Sharding->>Sharding: Route to BmmEPShardingInfo
Sharding->>Sharding: Shard selected_experts and final_scales per rank
Sharding->>Sharding: Insert all-reduce after MoE node
Sharding->>Graph: Apply tensor-parallel sharding
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Areas requiring extra attention:
Pre-merge checks and finishing touches❌ Failed checks (1 warning, 1 inconclusive)
✅ 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}
Contributor
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 (3)
tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py (1)
1591-1599: Potential issue with transform list routing.The
_transform_list_dictmapping at line 1598 addsBmmEPShardingInfoafterEPShardingInfoat line 1596. SinceBmmEPShardingInfois a subclass ofEPShardingInfo, and theadd()method usesisinstance()to find the appropriate list,BmmEPShardingInfoinstances will matchEPShardingInfofirst and be routed toep_transformsinstead ofstacked_ep_transforms.Consider reordering so subclasses come before parent classes:
self._transform_list_dict = { WeightShardingInfo: self.weight_sharding_transforms, BMMShardingInfo: self.bmm_transforms, + BmmEPShardingInfo: self.stacked_ep_transforms, EPShardingInfo: self.ep_transforms, ParameterUpdateInfo: self.parameter_update_transforms, - BmmEPShardingInfo: self.stacked_ep_transforms, }tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py (1)
1-10: Missing NVIDIA copyright header.As per coding guidelines, all TensorRT-LLM Open Source Software code files should contain an NVIDIA copyright header that includes the current year at the top.
tests/unittest/_torch/helpers.py (1)
1-3: Missing NVIDIA copyright header.As per coding guidelines, all TensorRT-LLM Open Source Software code files should contain an NVIDIA copyright header that includes the current year at the top.
🧹 Nitpick comments (3)
tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py (1)
167-169: Clarify test setup: overridingselected_expertsdiscards actual routing.Line 169 replaces the actual topk routing decisions with all 1s, routing every token to expert 1. Combined with line 168 pre-scaling the input, this creates a specific but potentially confusing test scenario.
Consider adding a comment explaining why this setup is intentional, or use the actual routing from
setup_bmm_moe_testfor more realistic coverage.with torch.inference_mode(): + # Pre-scale input and force all tokens to expert 1 to test + # uniform routing scenario with BMM-based MoE x = final_scales * x selected_experts = torch.ones_like(selected_experts)tests/unittest/_torch/helpers.py (1)
131-139: Document theapply_routing_on_inputparameter.The docstring is missing documentation for the
apply_routing_on_inputparameter. This is important since it has a different default value (True) compared toreference_moe_torch(False)."""Reference for stacked MoE (torch_moe_bmm) in TRT-LLM format. Args: x: (seq_len, hidden_size) selected_experts: (seq_len, topk) final_scales: (seq_len, topk) w3_w1_stacked_weight: (num_experts, 2*intermediate_size, hidden_size) - TRT-LLM format w2_stacked_weight: (num_experts, hidden_size, intermediate_size) - TRT-LLM format + apply_routing_on_input: If True, apply routing to INPUT (default for Llama4 pattern). + If False, apply routing to OUTPUT (alternative pattern). + """tensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py (1)
164-190: Consider simplifying dtype handling and ones_node creation.Two observations:
Dtype condition (line 168): The condition
weight_dtype != torch.float32casts when weights are NOT float32. However, this doesn't check if the input already matches the weight dtype, potentially adding unnecessary cast operations.Ones node overhead (lines 186-190): Creating a tensor of ones to prevent the kernel from applying routing adds runtime overhead. Consider adding a kernel flag to skip routing multiplication instead.
- if weight_dtype is not None and weight_dtype != torch.float32: + # Cast to weight dtype only if input dtype differs + if weight_dtype is not None and hidden_states.meta.get('tensor_meta', {}).dtype != weight_dtype: input_to_scale = graph.call_function(For the ones_node, this could be a future optimization if profiling shows it's a bottleneck.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (8)
tensorrt_llm/_torch/auto_deploy/config/default.yaml(1 hunks)tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py(6 hunks)tensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py(2 hunks)tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py(3 hunks)tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py(6 hunks)tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_ep_sharding.py(1 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py(3 hunks)tests/unittest/_torch/helpers.py(2 hunks)
🧰 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 in Python, even if only one class or function from a module is used (e.g., usefrom package.subpackage import fooand thenfoo.SomeClass()instead offrom package.subpackage.foo import SomeClass)
Python filenames should use snake_case (e.g.,some_file.py)
Python class names should use PascalCase (e.g.,class SomeClass)
Python function and method names should use snake_case (e.g.,def my_awesome_function():)
Python local variable names 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
Python comments 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
Python attributes and variables can be documented inline with type and description (e.g.,self.x = 5followed by"""<type>: Description of 'x'""")
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 specific errors possible instead of catching all exceptions
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 to implement the logic
Files:
tests/unittest/_torch/helpers.pytensorrt_llm/_torch/auto_deploy/transform/library/sharding.pytensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.pytests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.pytests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_ep_sharding.pytensorrt_llm/_torch/auto_deploy/utils/sharding_utils.pytensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
**/*.{cpp,h,cu,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code files should contain an NVIDIA copyright header that includes the current year at the top
Files:
tests/unittest/_torch/helpers.pytensorrt_llm/_torch/auto_deploy/transform/library/sharding.pytensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.pytests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.pytests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_ep_sharding.pytensorrt_llm/_torch/auto_deploy/utils/sharding_utils.pytensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
🧠 Learnings (8)
📓 Common learnings
Learnt from: djns99
Repo: NVIDIA/TensorRT-LLM PR: 6915
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu:4010-4012
Timestamp: 2025-08-14T23:23:27.449Z
Learning: For MOE (Mixture of Experts) code reviews in TensorRT-LLM, avoid repeatedly suggesting finalize fusion validation checks and safety assertions. The user djns99 has indicated these suggestions are repetitive and unwanted across multiple MOE-related changes.
Learnt from: jhaotingc
Repo: NVIDIA/TensorRT-LLM PR: 7856
File: cpp/tensorrt_llm/thop/fp8BlockScaleMoe.cpp:159-166
Timestamp: 2025-09-19T21:28:13.751Z
Learning: In TensorRT-LLM blockScaleMoe routing (cpp/tensorrt_llm/kernels/trtllmGenKernels/blockScaleMoe/runner.cu), the DeepSeek routing method performs reinterpret_cast<float*>(routingLogits) at line 89, which could cause issues if routing_logits are BF16. However, Qwen3-FP8 models use RenormalizeNaive routing method and are not affected by this dtype casting issue.
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.
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 allreduce implementation (cpp/tensorrt_llm/thop/allreduceOp.cpp), the goto pattern in runNCCLAllReduceDeviceFusion is intentionally used for future extensibility, allowing multiple switch cases to fallback to the default handler. While not aesthetically ideal, this pattern supports adding more fusion cases later that can reuse the same fallback logic.
📚 Learning: 2025-09-19T21:28:13.751Z
Learnt from: jhaotingc
Repo: NVIDIA/TensorRT-LLM PR: 7856
File: cpp/tensorrt_llm/thop/fp8BlockScaleMoe.cpp:159-166
Timestamp: 2025-09-19T21:28:13.751Z
Learning: In TensorRT-LLM blockScaleMoe routing (cpp/tensorrt_llm/kernels/trtllmGenKernels/blockScaleMoe/runner.cu), the DeepSeek routing method performs reinterpret_cast<float*>(routingLogits) at line 89, which could cause issues if routing_logits are BF16. However, Qwen3-FP8 models use RenormalizeNaive routing method and are not affected by this dtype casting issue.
Applied to files:
tests/unittest/_torch/helpers.pytensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py
📚 Learning: 2025-08-08T22:03:40.707Z
Learnt from: sklevtsov-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 3294
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu:1198-1209
Timestamp: 2025-08-08T22:03:40.707Z
Learning: In the CUTLASS MoE kernels (cpp/tensorrt_llm/cutlass_extensions), when `layout_info.fusion` is set to `TmaWarpSpecializedGroupedGemmInput::EpilogueFusion::FINALIZE`, the `router_scales` parameter must be non-null by design. The fused finalize kernel epilogue does not perform nullptr checks and requires valid router scales to function correctly. This is an implicit contract that callers must satisfy when enabling the FINALIZE fusion mode.
Applied to files:
tests/unittest/_torch/helpers.pytensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py
📚 Learning: 2025-08-14T23:23:27.449Z
Learnt from: djns99
Repo: NVIDIA/TensorRT-LLM PR: 6915
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu:4010-4012
Timestamp: 2025-08-14T23:23:27.449Z
Learning: For MOE (Mixture of Experts) code reviews in TensorRT-LLM, avoid repeatedly suggesting finalize fusion validation checks and safety assertions. The user djns99 has indicated these suggestions are repetitive and unwanted across multiple MOE-related changes.
Applied to files:
tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.pytensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
📚 Learning: 2025-08-21T02:39:12.009Z
Learnt from: djns99
Repo: NVIDIA/TensorRT-LLM PR: 7104
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu:1475-1480
Timestamp: 2025-08-21T02:39:12.009Z
Learning: The min latency mode functionality in TensorRT-LLM MOE kernels (cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_kernels.cu) is deprecated and no longer being maintained/updated, as confirmed by djns99. Bug reports and optimization suggestions for the computeStridesTmaWarpSpecializedLowLatencyKernel and related min latency code paths should be deprioritized.
Applied to files:
tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.pytensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
📚 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/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py
📚 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/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.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 allreduce implementation (cpp/tensorrt_llm/thop/allreduceOp.cpp), the goto pattern in runNCCLAllReduceDeviceFusion is intentionally used for future extensibility, allowing multiple switch cases to fallback to the default handler. While not aesthetically ideal, this pattern supports adding more fusion cases later that can reuse the same fallback logic.
Applied to files:
tensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
🧬 Code graph analysis (3)
tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py (2)
tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py (5)
BmmEPShardingInfo(1508-1518)add(1624-1654)from_node(660-665)from_node(1439-1444)EPShardingInfo(1429-1455)tensorrt_llm/_torch/auto_deploy/utils/node_utils.py (1)
is_op(198-221)
tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py (3)
tests/unittest/_torch/helpers.py (2)
reference_bmm_moe_torch(124-170)reference_moe_torch(96-121)tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py (1)
torch_moe_bmm(175-227)tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/trtllm_moe.py (1)
trtllm_moe_fused(7-54)
tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py (4)
tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py (1)
ShardingTransformConfig(65-94)tensorrt_llm/functional.py (1)
AllReduceStrategy(3876-3885)tests/unittest/_torch/auto_deploy/unit/multigpu/custom_ops/test_mxfp4_moe_ep.py (1)
_split_range_last_remainder(14-19)tensorrt_llm/_torch/auto_deploy/utils/node_utils.py (1)
is_op(198-221)
🪛 Ruff (0.14.6)
tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py
233-233: Unused function argument: selected_experts
(ARG001)
234-234: Unused function argument: routing_weights
(ARG001)
235-235: Unused function argument: w3_w1_stacked
(ARG001)
236-236: Unused function argument: w2_stacked
(ARG001)
237-237: Unused function argument: act_fn
(ARG001)
238-238: Unused function argument: apply_routing_on_input
(ARG001)
tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py
1274-1274: Avoid specifying long messages outside the exception class
(TRY003)
1293-1293: Avoid specifying long messages outside the exception class
(TRY003)
1511-1511: Unused method argument: gm
(ARG002)
tensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py
54-54: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
64-64: Avoid specifying long messages outside the exception class
(TRY003)
1059-1059: Unused method argument: cm
(ARG002)
1060-1060: Unused method argument: factory
(ARG002)
1061-1061: Unused method argument: shared_config
(ARG002)
1070-1070: zip() without an explicit strict= parameter
Add explicit value for parameter strict=
(B905)
1070-1070: Prefer itertools.pairwise() over zip() when iterating over successive pairs
Replace zip() with itertools.pairwise()
(RUF007)
⏰ 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 (28)
tensorrt_llm/_torch/auto_deploy/config/default.yaml (1)
31-32: LGTM!The new
match_bmm_moe_patterntransform entry is correctly placed in the pattern_matcher stage alongside the existing MoE pattern matchers. The ordering aftermatch_dense_moe_patternis appropriate.tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py (4)
29-84: LGTM!The
apply_routing_on_inputparameter is well-implemented with clear documentation. The conditional logic correctly handles both routing modes:
- INPUT-SIDE:
silu(input * routing_weight)- routing affects activation- OUTPUT-SIDE:
silu(input) * routing_weight- routing scales outputThe default value
Falsemaintains backward compatibility with existing callers.
174-227: LGTM!The new
torch_moe_bmmoperator correctly implements BMM-based MoE with stacked TRT-LLM format weights. The implementation:
- Properly extracts W3 (first half) and W1 (second half) from the fused gate_up tensor
- Handles dtype conversion for mixed-precision scenarios
- Delegates to
_template_moewith the configurable routing behaviorThe documentation clearly specifies the expected weight shapes and routing semantics.
230-240: LGTM!The fake implementation correctly returns
torch.empty_like(x)for shape inference during tracing. The static analysis warnings about unused arguments (ARG001) are expected for fake/stub implementations where the signature must match the real operator.
243-295: LGTM!The
torch_fused_moeimplementation correctly follows the SwiGLU computation pattern with TRT-LLM format weights. The docstring updates clearly document the expected weight shapes and format conversions.tensorrt_llm/_torch/auto_deploy/utils/sharding_utils.py (6)
1096-1153: LGTM!The
_slice_expert_dimfunction correctly handles bothNodeandTensorinputs with proper load hook registration for checkpoint compatibility. The edge case handling for empty module names at line 1127 is appropriate.
1156-1258: LGTM with observation.The
_slice_and_transpose_expert_dimfunction correctly handles the Llama4 to TRT-LLM format conversion with W1/W3 swapping and transpose operations.Minor observation: At lines 1204-1206, when
swap_gate_up=Truefor non-parameter nodes, the code logs a warning but continues without actually swapping. This is acceptable given the primary use case is parameter nodes, but worth noting for future maintainers.
1261-1274: LGTM!The
_get_dim0_from_arghelper correctly handles multiple input types (Tensor, get_attr Node, and meta val) with appropriate fallback logic.
1277-1357: LGTM!The
_insert_sharded_moe_bmmfunction correctly implements EP sharding for BMM-based MoE nodes with stacked weights. The implementation:
- Properly handles selected_experts and final_scales sharding
- Applies Llama4 to TRT-LLM format conversion during weight slicing
- Adds the required all_reduce for result aggregation
The pattern follows the existing
_insert_sharded_moeimplementation for consistency.
1508-1518: LGTM!The
BmmEPShardingInfoclass correctly extendsEPShardingInfowith appropriate validation and apply methods fortorch_moe_bmmnodes. The unusedgmargument invalidateis expected for interface consistency with the parent class.
1521-1527: LGTM!The EP sharding rules are correctly updated to include the mapping for
torch_moe_bmmtoBmmEPShardingInfo.tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py (2)
141-143: LGTM!The executor loop for
stacked_ep_transformsfollows the established pattern for processing sharding transforms and correctly incrementsnum_matches.
1060-1090: LGTM!The
detect_ep_shardfunction correctly handles the newtorch_moe_bmmop by:
- Including it in the MoE op whitelist
- Routing it to
BmmEPShardingInfo.from_nodefor proper sharding info creationThe conditional logic cleanly separates BMM-based MoE from other variants while maintaining the existing pattern for other MoE operations.
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_ep_sharding.py (1)
162-201: Verify attribute names in SharedConfig class.The suggested fix changes
sharding_configtosharding_transform_containerbut the attribute naming betweenstacked_ep_transformsvsep_transformsneeds clarification by cross-referencing the SharedConfig definition and line 132 of the same test file.tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py (2)
65-106: LGTM!The setup function correctly initializes test data for BMM-based MoE with topk=1 and TRT-LLM formatted stacked weights.
191-204: Verify routing weight application is not doubled.The input
xis pre-scaled byfinal_scalesat line 168, andreference_bmm_moe_torchis called withapply_routing_on_input=Truewhich scales again. Verify thattorch_moe_bmm,torch_moe_fused, andtrtllm_moe_fusedoperators all consistently apply or skip routing scaling internally; if they behave differently, the test tolerances and comparison assertions may mask operator divergence.tests/unittest/_torch/helpers.py (2)
96-121: LGTM!The
apply_routing_on_inputparameter addition correctly implements conditional routing—scaling inputs before MLP computation when True, or scaling outputs after when False.
140-170: LGTM!The implementation correctly handles TRT-LLM weight format with proper slicing and transposition, and the conditional routing logic mirrors
reference_moe_torch.tensorrt_llm/_torch/auto_deploy/transform/library/fused_moe.py (10)
21-92: LGTM!The function correctly stacks per-expert weights into fused tensors and registers them as parameters. The handling of both
gated_mlpandmlpstyles is appropriate.
204-257: LGTM!The unified handling of both standard MoE and Llama4 BMM MoE patterns is clean. The
fused_key_counterincrement placement after the if/else is correct.
740-760: LGTM!The config class and transform registration follow the established pattern. The docstring clearly states the topk=1 constraint.
762-827: LGTM!The
_find_gate_up_bmmstatic method correctly traces back through the BMM MoE pattern:final_bmm ← mul(up, silu(gate)) ← chunk ← first_bmm. The chunk validation (chunk.args[1] != 2) ensures we're matching the expected gate/up split.
829-891: LGTM!The
_find_input_and_routingmethod correctly validates the topk=1 constraint and traces back to find original input and routing information.
893-944: LGTM!The
_find_output_and_routing_flavormethod properly detects both INPUT-SIDE and OUTPUT-SIDE routing patterns by checking for multiplication after the sum operation.
1056-1061: Unused arguments are acceptable for interface compliance.The
cm,factory, andshared_configarguments are required by theBaseTransform._applyinterface. The static analysis warnings can be safely ignored.
1070-1076: Consideritertools.pairwise()for Python 3.10+.Static analysis suggests using
itertools.pairwise()for successive pairs. However, since the codebase targets Python 3.8+ per coding guidelines, the currentzip()approach is appropriate. Thestrict=parameter suggestion is also not critical here since both slices come from the same list.
1164-1192: LGTM!The fusion correctly creates the
torch_moe_bmmnode with extracted routing information and properly cleans up dead nodes after replacement.
1114-1157: Verify whether the 2-hop heuristic for scatter/sigmoid detection is sufficient for production patterns.The code allows up to 2 intermediate operations between scatter and sigmoid (specifically designed for
to.dtypeconversions), but whether this covers all real graph structures in MoE model tracing remains unverified. Review the actual patterns produced during model compilation and check if any exceed this limit.
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #26301 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #26301 [ run ] completed with state |
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #26325 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #26325 [ run ] completed with state |
…ring Signed-off-by: Tal Cherckez <[email protected]>
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #26414 [ run ] triggered by Bot. Commit: |
nzmora-nvidia
approved these changes
Dec 1, 2025
tests/unittest/_torch/auto_deploy/unit/singlegpu/custom_ops/test_ad_moe_op.py
Outdated
Show resolved
Hide resolved
galagam
approved these changes
Dec 1, 2025
Collaborator
galagam
left a comment
galagam
left a comment
There was a problem hiding this comment.
Looks good, added a minor comment.
…st_ad_moe_op.py Co-authored-by: Neta Zmora <[email protected]> Signed-off-by: tcherckez-nvidia <[email protected]>
Signed-off-by: Tal Cherckez <[email protected]>
Collaborator
|
PR_Github #26414 [ run ] completed with state |
nvchenghaoz
approved these changes
Dec 1, 2025
tensorrt_llm/_torch/auto_deploy/custom_ops/fused_moe/torch_moe.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Tal Cherckez <[email protected]>
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #26602 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #26602 [ run ] completed with state |
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #26746 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #26746 [ run ] completed with state |
1 task
MinaHuai
pushed a commit
to davidmlw/TensorRT-LLM
that referenced
this pull request
Dec 10, 2025
…VIDIA#8779) The performance results of some kernels could be easily affected by the warm/cold L2 cache status. To achieve more precise profiling results, the L2 cache is cleared for every execution by the circular buffer method for better benchmarking during autotuning. Signed-off-by: Yukun He <[email protected]> [None][infra] Waive failed cases for main branch on 11/25 (NVIDIA#9429) Signed-off-by: qqiao <[email protected]> [NVIDIA#8391][chore] test_perf.py to lock clocks read from gpu_configs.yml instead of max freq (NVIDIA#9409) Signed-off-by: Eran Geva <[email protected]> [None][ci] Move more test stages to use OCI machines (NVIDIA#9395) Signed-off-by: Yanchao Lu <[email protected]> Co-authored-by: Matt Lefebvre <[email protected]> [None][feat] Improve TRTLLM MoE in small hidden size throughput cases (NVIDIA#9377) Signed-off-by: Anthony Chang <[email protected]> [https://nvbugs/5537996][fix] Let KV cache manager block initialization be aware whether it is doing a dry run or not (NVIDIA#9093) Before this commit, the kv cache manager does the same regardless, which causes a mis-calculation in free memory available to allocate for the KV cache manager, hence causing a crash. This commit fixes this by letting KV cache manager initialization be aware whether it is doing the dry run or not. If it is a dry run, use the max_tokens setting that is already pre-calculated and filled into kv_cache_config.max_tokens. Signed-off-by: eopXD <[email protected]> [https://nvbugs/5667922][fix] Update long context evaluation config (NVIDIA#9426) Signed-off-by: mni <[email protected]> [None][fix] Mitigate test timeout issues (NVIDIA#9445) Signed-off-by: Shixiaowei02 <[email protected]> [None][chore] Fix trtllm-eval for PyTorchLLM (NVIDIA#9427) Signed-off-by: Fanrong Li <[email protected]> [None][feat] Add a parser to layer-wise benchmarks (NVIDIA#9440) Signed-off-by: Tailing Yuan <[email protected]> [None][feat] Support custom chat template for tool calling (NVIDIA#9297) Signed-off-by: Pengyun Lin <[email protected]> [TRTLLM-8160][feat] Add draft token tree runtime on CDL (NVIDIA#8586) Signed-off-by: Yue Weng <[email protected]> [None][ci] waive a test (NVIDIA#9458) Signed-off-by: Yan Chunwei <[email protected]> [https://nvbugs/5680905][fix] Relax the MMLU accuracy requirement for DS-v3.2 (NVIDIA#9439) Signed-off-by: Fanrong Li <[email protected]> [TRTLLM-8376][feat] top-p optimization (removes redundant softmax) (NVIDIA#9411) Signed-off-by: ixlmar <[email protected]> [TRTLLM-9490][feat] use FlashInfer's top_k_sampling_from_probs (NVIDIA#9457) Signed-off-by: ixlmar <[email protected]> [https://nvbugs/5647400] [fix] Enlarged the AllReduce workspace size to 64MB. Added AllReduce strategy to AD config. (NVIDIA#9145) Signed-off-by: Eran Geva <[email protected]> [TRTLLM-909][feat] Overlap context chunks in pipeline parallel mode (NVIDIA#9308) Signed-off-by: Robin Kobus <[email protected]> [None][chore] AutoDeploy add multi stream moe pass to default.yaml (NVIDIA#9430) Signed-off-by: Suyog Gupta <[email protected]> [https://nvbugs/5685143][fix] avoid cudaFree overlap with cuda graph (NVIDIA#9438) Signed-off-by: Chuang Zhu <[email protected]> [None][chore] Bump version to 1.2.0rc5 (NVIDIA#9455) Signed-off-by: Yiqing Yan <[email protected]> [TRTLLM-8936][test] Add disagg and wideep multi-node multi-gpu test cases (NVIDIA#9356) Signed-off-by: FredricZ-2007 <[email protected]> [None][ci] move some slow test cases of DGX-B200 to post merge (NVIDIA#9467) Signed-off-by: junq <[email protected]> [TRTLLM-9293][feat] Enable partial weight loading to support streaming update weights (NVIDIA#9224) Signed-off-by: shuyix <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [TRTLLM-9264][fix] Add accuracy/unit tests/doc for phi4mm (NVIDIA#9246) Signed-off-by: Wanli Jiang <[email protected]> [https://nvbugs/5580099][fix] Cherry pick IMA issue fix from release/1.1 (NVIDIA#9032) Signed-off-by: Junyi Xu <[email protected]> [None][chore] Upgrade CuteDSL to 4.3.0 (NVIDIA#9444) Signed-off-by: Enwei Zhu <[email protected]> [None][feat] Support MLA chunked prefill for DeepSeek V3.2 model (NVIDIA#9376) Signed-off-by: Chang Liu (Enterprise Products) <[email protected]> [None][feat] Add environment variable to force spec-dec number of accepted tokens (NVIDIA#9371) Signed-off-by: Aurelien Chartier <[email protected]> [None][infra] Update allowed list 2025.11.25 (NVIDIA#9468) Signed-off-by: Yuanjing Xue <[email protected]> [None][infra] Fail the pipeline when slurm ssh dropped (NVIDIA#9157) Signed-off-by: Yuanjing Xue <[email protected]> [None][feat] AutoDeploy: Remove redundant copies in mamba layers (NVIDIA#9461) Signed-off-by: Chenghao Zhang <[email protected]> Co-authored-by: Suyog Gupta <[email protected]> [None][feat] AutoDeploy: Add A_log fusion for Mamba layers (NVIDIA#9422) Signed-off-by: Chenghao Zhang <[email protected]> [None][ci] Waive blackwell test on spec gate. (NVIDIA#9502) Signed-off-by: Zheyu Fu <[email protected]> [https://nvbugs/5608930][fix] Fix a typo (NVIDIA#9487) Signed-off-by: Shixiaowei02 <[email protected]> [NVIDIA#9463][feat] Add revision option to trtllm commands (NVIDIA#9498) Signed-off-by: Aurelien Chartier <[email protected]> [TRTLLM-9085][doc] fix math formula rendering issues (NVIDIA#9481) Signed-off-by: junq <[email protected]> [None][chore] update comments in llm_args.py (NVIDIA#9472) Signed-off-by: junq <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [https://nvbugs/5680310][fix] Fix ctx only timed out test (NVIDIA#9410) Signed-off-by: Patrice Castonguay <[email protected]> [https://nvbugs/5547414][fix] enable case after using local cache model (NVIDIA#9473) Signed-off-by: Hui Gao <[email protected]> [None][fix] Replace PYTORCH_CUDA_ALLOC_CONF with PYTORCH_ALLOC_CONF to fix deprecation warning (NVIDIA#9294) Signed-off-by: Jiagan Cheng <[email protected]> [https://nvbugs/5698581][fix] Init draft tokens for CUDA graph dummy request (NVIDIA#9505) Signed-off-by: ziyixiong-nv <[email protected]> [None][infra] Waive failed case in pre-merge on 11/27 (NVIDIA#9507) Signed-off-by: qqiao <[email protected]> [TRTLLM-9513][docs] Qwen3 deployment guide (NVIDIA#9488) Signed-off-by: Lanyu Liao <[email protected]> Co-authored-by: Lanyu Liao <[email protected]> [None][chore] revert batch_size=1 to prevent timeout and lower accuracy reference by 0.12% as a WAR (NVIDIA#9447) Signed-off-by: Lizhi Zhou <[email protected]> Co-authored-by: Shi Xiaowei <[email protected]> [TRTLLM-9279][infra] Use flexcache for gh200 nodes since they locate in Austin (NVIDIA#9405) Signed-off-by: qqiao <[email protected]> Signed-off-by: Emma Qiao <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> [cherry-pick][https://nvbugs/5670793][fix] Solve trtllm-serve launch_disaggregated issue (NVIDIA#9346) Signed-off-by: xxi <[email protected]> [None][infra] Fix Slurm job script (NVIDIA#9508) Signed-off-by: Yuanjing Xue <[email protected]> [None][fix] change allreduce workspace dtype to torch.int64 to avoid overflow (NVIDIA#9479) Signed-off-by: Zhenhuan Chen <[email protected]> [None][feat] add qwen3-next CI test of accuracy on BF16 and NVFP4 (NVIDIA#9330) Signed-off-by: jiant <[email protected]> [None][fix] fix TP support for DeepSeek-V3.2 on hopper (NVIDIA#9484) Signed-off-by: Fanrong Li <[email protected]> [TRTLLM-9389][chore] Refactor AlltoallMethodType. (NVIDIA#9388) Signed-off-by: Bo Li <[email protected]> [https://nvbugs/5674665][chore] Add test coverage for https://nvbugspro.nvidia.com/bug/5674665 (NVIDIA#9518) Signed-off-by: eopXD <[email protected]> [TRTLLM-7288][infra] Download merged waive list in slurm script (NVIDIA#8999) Signed-off-by: Yiqing Yan <[email protected]> Signed-off-by: Yanchao Lu <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> [https://nvbugs/5687820][fix] Remove self.abort() in DetokenizedGenerationResult (NVIDIA#9449) Signed-off-by: Enwei Zhu <[email protected]> [NVIDIA#9150][feat] AutoDeploy Nemotron-Flash support (NVIDIA#9504) Signed-off-by: Lucas Liebenwein <[email protected]> [None] [chore] Update to cutlass 4.3 (NVIDIA#8637) Signed-off-by: Kaiyu Xie <[email protected]> [https://nvbugs/5637037][chore] Update waive lists. (NVIDIA#9386) Signed-off-by: Bo Li <[email protected]> Signed-off-by: Enwei Zhu <[email protected]> Co-authored-by: Enwei Zhu <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [TRTLLM-8970][infra] Fix generate report when has isolation test result (NVIDIA#8861) Signed-off-by: qqiao <[email protected]> Signed-off-by: Emma Qiao <[email protected]> [https://nvbugs/5685015][fix] Update invalid max_token test (NVIDIA#9435) Signed-off-by: Junyi Xu <[email protected]> [None][fix] Fix on-disk cache and revise logger/statistics for AutoTuner. (NVIDIA#9211) Signed-off-by: Yukun He <[email protected]> [https://nvbugs/5689658][test] Fix gpu lock issue running on cluster (NVIDIA#9441) Signed-off-by: yufeiwu <[email protected]> [None][chore] add spec_decoding configs in perf benchmark scripts and fix typos (NVIDIA#9533) Signed-off-by: Lanyu Liao <[email protected]> Co-authored-by: Lanyu Liao <[email protected]> [None][fix] Remove FP8 K/V buffer from TRTLLM sparse MLA attention kernel (NVIDIA#9529) Signed-off-by: Chang Liu (Enterprise Products) <[email protected]> [None] [chore] Enhancements and clean up to slurm scripts (NVIDIA#9493) Signed-off-by: Kaiyu Xie <[email protected]> [None][chore] Revert "[None][fix] change allreduce workspace dtype to torch.int64 t… (NVIDIA#9538) Signed-off-by: Zhenhuan Chen <[email protected]> [None][infra] Waive failed cases for main branch on 11/28 (NVIDIA#9539) Signed-off-by: qqiao <[email protected]> [None][fix] Pass checkpoint_format to create_input_processor (NVIDIA#9521) Signed-off-by: Robin Kobus <[email protected]> [TRTLLM-9541][infra] Use artifactory mirror for download.pytorch.org (NVIDIA#9477) Signed-off-by: ZhanruiSunCh <[email protected]> Signed-off-by: Zhanrui Sun <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> [TRTLLM-9488][feat] add 'disable_flashinfer_sampling' config option (NVIDIA#9454) Signed-off-by: ixlmar <[email protected]> [None][infra] Waive failed case in pre-merge on 11/28 (NVIDIA#9537) Signed-off-by: Wangshanshan <[email protected]> [None][perf] Helix: improve all-to-all perf for large CP size (NVIDIA#9494) Signed-off-by: Matthias Jouanneaux <[email protected]> Signed-off-by: Zheyu Fu <[email protected]> Co-authored-by: Zheyu Fu <[email protected]> [None][feat] support for more accurate AR calculation (NVIDIA#9323) Signed-off-by: binghanc <[email protected]> [TRTLLM-9488][fix] llmapi references (NVIDIA#9547) Signed-off-by: ixlmar <[email protected]> [NVIDIA#8948][feat] Support custom sharding config (NVIDIA#9143) Signed-off-by: greg-kwasniewski1 <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None][chore] Weekly mass integration of release/1.1 -- rebase (NVIDIA#9522) Signed-off-by: yunruis <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Wangshanshan <[email protected]> Signed-off-by: qgai <[email protected]> Signed-off-by: Balaram Buddharaju <[email protected]> Signed-off-by: Yan Chunwei <[email protected]> Signed-off-by: Junyi Xu <[email protected]> Signed-off-by: Simeng Liu <[email protected]> Signed-off-by: nv-guomingz <[email protected]> Signed-off-by: Jin Li <[email protected]> Signed-off-by: Ivy Zhang <[email protected]> Signed-off-by: Vincent Zhang <[email protected]> Signed-off-by: peaceh <[email protected]> Signed-off-by: Michal Guzek <[email protected]> Signed-off-by: Michal Guzek <[email protected]> Signed-off-by: Chang Liu (Enterprise Products) <[email protected]> Signed-off-by: leslie-fang25 <[email protected]> Signed-off-by: Shunkang <[email protected]> Signed-off-by: junq <[email protected]> Co-authored-by: yunruis <[email protected]> Co-authored-by: sunnyqgg <[email protected]> Co-authored-by: brb-nv <[email protected]> Co-authored-by: Yan Chunwei <[email protected]> Co-authored-by: JunyiXu-nv <[email protected]> Co-authored-by: Simeng Liu <[email protected]> Co-authored-by: Guoming Zhang <[email protected]> Co-authored-by: Jin Li <[email protected]> Co-authored-by: Ivy Zhang <[email protected]> Co-authored-by: Vincent Zhang <[email protected]> Co-authored-by: peaceh-nv <[email protected]> Co-authored-by: Michal Guzek <[email protected]> Co-authored-by: Chang Liu <[email protected]> Co-authored-by: Leslie Fang <[email protected]> Co-authored-by: Shunkangz <[email protected]> Co-authored-by: Shunkang <[email protected]> Co-authored-by: QI JUN <[email protected]> [TRTLLM-5971][feat] Integrate helix parallelism (NVIDIA#9342) Signed-off-by: Balaram Buddharaju <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None][infra] - Request idle time exemption for OCI jobs (NVIDIA#9528) Signed-off-by: Yanchao Lu <[email protected]> [None][infra] Wiave failed tests for main branch on 11/30 (NVIDIA#9555) Signed-off-by: qqiao <[email protected]> [None][fix] Fix port conflict in disagg tests (NVIDIA#9474) Signed-off-by: Junyi Xu <[email protected]> [None][ci] Split H100_PCIe-PyTorch-Post-Merge test stage (NVIDIA#9558) Signed-off-by: Yanchao Lu <[email protected]> [None][ci] Split H100_PCIe-PyTorch-Post-Merge test stage (NVIDIA#9559) Signed-off-by: Yanchao Lu <[email protected]> [TRTLLM-8958][feat] and [TRTLLM-8960]: create ConfigurableMoE and support TRTLLMGenFusedMoE as backend (NVIDIA#9486) [None] [feat] Optimize the algorithm part of RocketKV (NVIDIA#9333) Signed-off-by: yuhangh <[email protected]> [https://nvbugs/5690172][fix] Fix Qwen3-235B ATP accuracy issue with PDL (NVIDIA#9530) Signed-off-by: Enwei Zhu <[email protected]> [TRTLLM-6222][feat] Extend cute_dsl_nvfp4_gemm to sm103. (NVIDIA#9543) Signed-off-by: Mindy Li <[email protected]> [None][fix] Correct virtual memory allocation alignment (NVIDIA#9491) Signed-off-by: Yuan Tong <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [https://nvbugs/5684703][fix] Unwaive disagg guided decoding test (NVIDIA#9466) Signed-off-by: Enwei Zhu <[email protected]> [https://nvbugs/5503479][fix] Temporarily lower reference accuracy to stabilize CI (NVIDIA#9398) Signed-off-by: Pengbo Wang <[email protected]> [None][chore] remove qwen3-next accuracy tests (NVIDIA#9534) Signed-off-by: jiant <[email protected]> [None][doc] fix mtp.py typo (NVIDIA#9307) Signed-off-by: liugaoji <[email protected]> [None][feat] add chat template kwargs support to longbench-v2 (NVIDIA#9544) Signed-off-by: Fanrong Li <[email protected]> [NVIDIA#9496][fix] AutoDeploy: remove auto-tuner from nvfp4_gemm forward (NVIDIA#9497) Signed-off-by: Neta Zmora <[email protected]> [None][fix] Replace hash method with unique_id for cutedsl MoE runners. (NVIDIA#9569) Signed-off-by: Yukun He <[email protected]> [None][chore] refactor disaggregated scripts to use named arguments (NVIDIA#9581) Signed-off-by: Zhenhuan Chen <[email protected]> [TRTLLM-6222][feat] Several perf opt for cuteDSL nvf4 gemm (NVIDIA#9428) Signed-off-by: Yuhan Li <[email protected]> [None][chore] reduce the layers of the `devel` docker image (NVIDIA#9077) Signed-off-by: Martin Marciniszyn Mehringer <[email protected]> [https://nvbugs/5651854][infra] Enable perf metrics during accuracy testing (NVIDIA#9140) [None][fix] Skip Allreduce init for Attention DP (NVIDIA#9542) Signed-off-by: Enwei Zhu <[email protected]> [None][test] [None][test] Waive main branch test failures 12/1 (NVIDIA#9566) Signed-off-by: Yanchao Lu <[email protected]> [None][ci] Minor change for Slurm scripts (NVIDIA#9561) Signed-off-by: Yanchao Lu <[email protected]> [TRTLLM-6768][infra] Fix params for not updating github status (NVIDIA#6747) Signed-off-by: Yiqing Yan <[email protected]> [None][infra] Update the pytest options after MI (NVIDIA#9579) Signed-off-by: qqiao <[email protected]> [TRTLLM-6756][feat] Add Beam Search to TorchSampler (NVIDIA#8509) Signed-off-by: Stefan Niebler <[email protected]> [None][chore] Defer exposing context parallel configs (NVIDIA#9552) Signed-off-by: Balaram Buddharaju <[email protected]> [TRTC-1943][feat] Env vars override support in LLM API (NVIDIA#9104) Signed-off-by: Venky Ganesh <[email protected]> [None][feat] AutoDeploy: Use the router gemm op for nemotron MOE (NVIDIA#9500) Signed-off-by: Chenghao Zhang <[email protected]> [NVIDIA#9198][feat] Refactor dist ops in AutoDeploy (NVIDIA#9301) Signed-off-by: Eran Geva <[email protected]> [None][fix] Prevent YAML partial kv_cache_config from incorrectly overriding the complete kv_cache_config (NVIDIA#9262) Signed-off-by: Yuening Li <[email protected]> [TRTLLM-9085][doc] fix math formula rendering issues in github (NVIDIA#9605) Signed-off-by: junq <[email protected]> [None][feat] Unify nvfp4 gemm backend (NVIDIA#8963) Signed-off-by: Shijie Wang <[email protected]> Signed-off-by: Yukun He <[email protected]> Signed-off-by: Shijie <[email protected]> Co-authored-by: Yukun He <[email protected]> [None][feat] Add support for KVCache reuse for DSv32 (NVIDIA#9383) Signed-off-by: Iman Tabrizian <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None][chroe] Polish qwen3-next modeling code. (NVIDIA#8902) Signed-off-by: nv-guomingz <[email protected]> [https://nvbugs/5703953][fix] Use random port for disagg tests (NVIDIA#9582) Signed-off-by: Junyi Xu <[email protected]> [None][fix] Waive gb200 (NVIDIA#9580) Signed-off-by: Xin He (SW-GPU) <[email protected]> [FMDL-1328][feat] Add support for nano-v3 and super-v3 with pytorch backend (NVIDIA#9261) Signed-off-by: Wanli Jiang <[email protected]> [https://nvbugs/5582091][test] increase warmup times in testing for multi-gpu cases (NVIDIA#9578) Signed-off-by: Ruodi Lu <[email protected]> Co-authored-by: Ruodi Lu <[email protected]> [None][chore] Add failed cases into waives.txt (NVIDIA#9588) Signed-off-by: xinhe-nv <[email protected]> [https://nvbugs/5702793][fix] Fix uncontiguous tensor view (NVIDIA#9576) Signed-off-by: shuyix <[email protected]> [None][infra] Waive failed cases for main branch (NVIDIA#9615) Signed-off-by: qqiao <[email protected]> [TRTLLM-9488][feat] use FlashInfer.sampling by default (NVIDIA#9545) Signed-off-by: ixlmar <[email protected]> [None][infra] Update allowlist 2025/12/01 (NVIDIA#9616) Signed-off-by: Yuanjing Xue <[email protected]> [None][infra] Remove an invalid test name in waives.txt (NVIDIA#9620) Signed-off-by: qqiao <[email protected]> Lock the gpu clocks in L0 perf tests (NVIDIA#9585) Signed-off-by: Eran Geva <[email protected]> [TRTLLM-9466][test] Evaluate helix parallelism with DSV3 Lite (NVIDIA#9597) Signed-off-by: Balaram Buddharaju <[email protected]> [None][fix] Extract GPU count from single-node stage names (NVIDIA#9599) Signed-off-by: Chang Liu (Enterprise Products) <[email protected]> [https://nvbugs/5667774][fix] Refine Piecewise Cuda Graph Condition for DP (NVIDIA#9393) Signed-off-by: Jin Li <[email protected]> [TRTLLM-9144][fix] enhance RPC robustness (NVIDIA#8711) Signed-off-by: Superjomn <[email protected]> Signed-off-by: Erin Ho <[email protected]> Signed-off-by: Yan Chunwei <[email protected]> Co-authored-by: Erin Ho <[email protected]> [https://nvbugs/5627710][fix] Fix synchronization bugs in KvCacheTransferManager that can cause corrupted blocks (NVIDIA#9056) Signed-off-by: thorjohnsen <[email protected]> Signed-off-by: Thor Johnsen <[email protected]> Co-authored-by: Iman Tabrizian <[email protected]> Co-authored-by: Robin Kobus <[email protected]> [TRTLLM-8980][test] Clean up spec dec tests in test_llm_api_pytorch (NVIDIA#8889) Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [NVIDIA#9150][feat] Add code for nano v3 to custom implementation in AD (NVIDIA#9465) * Why? We would like to show an alternative to monkey-patching in AutoDeploy. * What? This commit builds on the existing custom model implementation for NemotronH and adds the bits relevant for MoE layers. Part of NVIDIA#9150. Signed-off-by: William Zhang <[email protected]> [NVIDIA#9150][feat] AutoDeploy: reviewer comments for NVIDIA#9150 (NVIDIA#9527) Signed-off-by: Lucas Liebenwein <[email protected]> [https://nvbugs/5651854][fix] Fix dist-serving perf by clearing CPU affinity (NVIDIA#9549) Signed-off-by: Shixiaowei02 <[email protected]> [NVIDIA#9550][feat] AutoDeploy: Add NVFP4 Cutlass MoE kernels (NVIDIA#9551) Signed-off-by: Neta Zmora <[email protected]> [https://nvbugs/5688388][fix] fix: Reducing num request in disagg test to speed up (NVIDIA#9598) Signed-off-by: Patrice Castonguay <[email protected]> [TRTLLM-8946][feat] Improved heuristics to detect shardable regions (NVIDIA#9200) Signed-off-by: Lucas Liebenwein <[email protected]> Signed-off-by: greg-kwasniewski1 <[email protected]> Co-authored-by: Lucas Liebenwein <[email protected]> [NVIDIA#9632][feat] Support EXTRA_WHEEL_BUILD_ARGS during wheel build (NVIDIA#9633) Signed-off-by: Yu Chi Li <[email protected]> [None][chore] Waive test failing on pre-merge (NVIDIA#9638) Signed-off-by: Balaram Buddharaju <[email protected]> [None][chore] Remove traceback dump for multimodal input processor (NVIDIA#9634) Signed-off-by: Chang Liu (Enterprise Products) <[email protected]> [None][chore] Fix trtllm-eval and move GroupedGemmInputsHelper (NVIDIA#9612) Signed-off-by: Enwei Zhu <[email protected]> [https://nvbugs/5698434][fix] Use separate weight mapper for draft (NVIDIA#9607) Signed-off-by: Anurag Mukkara <[email protected]> [TRTLLM-7101][infra] Reuse passed tests (NVIDIA#6894) Signed-off-by: Yiqing Yan <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> [None][test] Remove duplicate test cases (NVIDIA#9623) Signed-off-by: yufeiwu <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None][feat] Add RocketKV usage doc and e2e accuracy test on LongBenchV2 (NVIDIA#9572) Signed-off-by: yuhangh <[email protected]> [TRTLLM-9242][doc] Add examples showcasing openai compatible APIs (NVIDIA#9520) Signed-off-by: Junyi Xu <[email protected]> [None][chore] AutoDeploy update cuda stream manager for multi-device (NVIDIA#9575) Signed-off-by: Suyog Gupta <[email protected]> [TRTLLM-9391][chore] Automatically estimate required workspace. (NVIDIA#9535) Signed-off-by: Bo Li <[email protected]> [https://nvbugs/5708475][fix] Fix e2e eval accuracy for helix parallelism (NVIDIA#9647) Signed-off-by: Balaram Buddharaju <[email protected]> [https://nvbugs/5561153][test] Fix log error for perf test (NVIDIA#9622) Signed-off-by: FredricZ-2007 <[email protected]> [TRTLLM-8241][feat] Aliasing to comply to LlmArgs (NVIDIA#9586) Signed-off-by: Pengyun Lin <[email protected]> [None][chore] Add failed cases into waives.txt (NVIDIA#9593) Signed-off-by: Jie Li <[email protected]> Co-authored-by: Jie Li <[email protected]> [TRTLLM-6842][feat] Support Response API for general purpose (NVIDIA#9392) Signed-off-by: Junyi Xu <[email protected]> [None][test] Update Qwen3-next accuracy testing by setting the cuda … (NVIDIA#9613) Signed-off-by: nv-guomingz <[email protected]> [None][feat] update trtllm-gen nvfp4 kernels with better performance (NVIDIA#9510) Signed-off-by: Perkz Zheng <[email protected]> [None][doc] Replace the tensorrt icon with torch icon on overview.md (NVIDIA#9644) Signed-off-by: nv-guomingz <[email protected]> [https://nvbugs/5705197][chore] Unwaive timeout disagg tests (NVIDIA#9637) Signed-off-by: Patrice Castonguay <[email protected]> [https://nvbugs/5552132][fix] Enable LoRa for GPT OSS Torch (NVIDIA#8253) Signed-off-by: Michal Guzek <[email protected]> [None][fix] Fix wide ep MoE error (NVIDIA#9642) Signed-off-by: Iman Tabrizian <[email protected]> [https://nvbugs/5702795][fix] Remove the warning message for aten.log. (NVIDIA#9665) Signed-off-by: nv-guomingz <[email protected]> [https://nvbugs/5693853][fix] Fix error handling when querying machin… (NVIDIA#9483) Signed-off-by: Gal Hubara Agam <[email protected]> [OMNIML-2932] [feat] nvfp4 awq support (NVIDIA#8698) Signed-off-by: weimingc <[email protected]> [NVIDIA#9643][fix] AutoDeploy: fix nano sharding config (NVIDIA#9668) Signed-off-by: Lucas Liebenwein <[email protected]> [NVIDIA#9147][feat] AutoDeploy: Draft Target Speculative Decoding (NVIDIA#9275) Signed-off-by: Govind Ramnarayan <[email protected]> [None][feat] Update Qwen3CodeToolParser to align tool-calling parameters (NVIDIA#9540) Signed-off-by: Wanli Jiang <[email protected]> [TRTLLM-7181][infra] Generate test results when pytest timeout happens (NVIDIA#9396) Signed-off-by: Yiqing Yan <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [TRTLLM-9522][fix] restore `trtllm-serve mm_embedding_serve` (NVIDIA#9669) [TRTLLM-5093][infra] Write env variables to a file in the interactive debug session (NVIDIA#6792) Signed-off-by: Yiqing Yan <[email protected]> [None][fix] fix error when processing batches containing both text and mm data (NVIDIA#8381) Signed-off-by: Nekofish-L <[email protected]> [TRTLLM-7073][feat] Support torch compile for PP for Llama and DeepSeekV3 (NVIDIA#7838) Signed-off-by: Jin Li <[email protected]> [None][feat] Add weights initialization and context phase parser to layer-wise benchmarks (NVIDIA#9667) Signed-off-by: Tailing Yuan <[email protected]> [TRTLLM-8274][feat] Check if executor is shutdown in /health entrypoint (NVIDIA#9057) Signed-off-by: Junyi Xu <[email protected]> [NVIDIA#8733][feat] Add Llama4 MoE handling to AutoDeploy (NVIDIA#9556) Signed-off-by: Tal Cherckez <[email protected]> Signed-off-by: tcherckez-nvidia <[email protected]> Co-authored-by: Neta Zmora <[email protected]> [None][ci] unwaive tests (NVIDIA#9651) Signed-off-by: Yan Chunwei <[email protected]> [None][feat] Add NIXL-LIBFABRIC support (NVIDIA#9225) Signed-off-by: Yoray Zack <[email protected]> Signed-off-by: zackyoray <[email protected]> [None][test] rename wide ep and disagg metric name in perf test (NVIDIA#9704) Signed-off-by: Ruodi Lu <[email protected]> Co-authored-by: Ruodi Lu <[email protected]> [https://nvbugs/5467531][fix] Unwaive fused_moe all to all test with … (NVIDIA#9617) Signed-off-by: Jin Li <[email protected]> [None][fix] Recover TRTLLM MoE Perf for DEP (NVIDIA#9562) Signed-off-by: Anthony Chang <[email protected]> [None][chore] Add failed cases into waives.txt (NVIDIA#9662) Signed-off-by: Xin He (SW-GPU) <[email protected]> Signed-off-by: xinhe-nv <[email protected]> Signed-off-by: Yanchao Lu <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> [None][fix] Fix TLLM_SPEC_DECODE_FORCE_NUM_ACCEPTED_TOKENS for MTP/EAGLE (NVIDIA#9608) Signed-off-by: Aurelien Chartier <[email protected]> [None][infra] Add container notices and documentation (NVIDIA#9185) Signed-off-by: Parker Drake <[email protected]> [TRTLLM-5312][infra] Add triton trigger rules (NVIDIA#6440) Signed-off-by: Yiqing Yan <[email protected]> [None][doc] Add feature docs for helix parallelism (NVIDIA#9684) Signed-off-by: Balaram Buddharaju <[email protected]> [TRTLLM-9579][infra] Set mergeWaiveList stage UNSTABLE when there is any issue (NVIDIA#9692) Signed-off-by: Yiqing Yan <[email protected]> [None][doc] Added line about partial reuse (NVIDIA#7846) Signed-off-by: thorjohnsen <[email protected]> [TRTLLM-8920][feat] decouple disagg service from fastapi (NVIDIA#8714) Signed-off-by: Lizhi Zhou <[email protected]> [https://nvbugs/5633340][fix] start disagg workers and servers on free ports (NVIDIA#9694) Signed-off-by: Lizhi Zhou <[email protected]> [TRTLLM-9562] [doc] Add Deployment Guide for Kimi K2 Thinking on TensorRT LLM - Blackwell (NVIDIA#9711) Signed-off-by: Kaiyu Xie <[email protected]> [NVIDIA#9602][feat] AutoDeploy: Support TRTLLM Sampler (NVIDIA#9641) Signed-off-by: Govind Ramnarayan <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None] [tests] Unwaive EPLB tests (NVIDIA#9625) Signed-off-by: Kaiyu Xie <[email protected]> [https://nvbugs/5518713][test] Refactor core test lists by merging with llm_perf_cluster.yml (NVIDIA#9714) Signed-off-by: yufeiwu <[email protected]> [TRTLLM-7136][feat] Update load_weights method to include mapping parameter in checkpoint loaders (NVIDIA#9583) Signed-off-by: Robin Kobus <[email protected]> [None][refactor] Improve request processing function in sampler (NVIDIA#9671) Signed-off-by: Robin Kobus <[email protected]> [https://nvbugs/5670672][fix] Fix flaky KV connector tests (NVIDIA#9676) Signed-off-by: jthomson04 <[email protected]> [None][infra] Update allowed list 20251204 (NVIDIA#9718) Signed-off-by: Yuanjing Xue <[email protected]> [None][feat] AutoDeploy: Perf optimization for Attention and rmsnorm (NVIDIA#9719) Signed-off-by: Chenghao Zhang <[email protected]> [None][chore] Waive flakey disagg tests (NVIDIA#9749) Signed-off-by: Mike Iovine <[email protected]> [https://nvbugs/5601682][fix] Fix cacheTransceiver hang (NVIDIA#9311) Signed-off-by: Iman Tabrizian <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9199][docs] KV Connector Docs (NVIDIA#9325) Signed-off-by: jthomson04 <[email protected]> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9160][doc] add doc to llm_runtime.py (NVIDIA#9482) Signed-off-by: Yan Chunwei <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [None][doc] VDR 1.0 trtllm-serve doc enhancement (NVIDIA#9443) Signed-off-by: Pengyun Lin <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9086][doc] Clean up TODOs in documentation (NVIDIA#9292) Signed-off-by: junq <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9157][doc] Guided decoding doc improvement (NVIDIA#9359) Signed-off-by: Enwei Zhu <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [None][infra] Updated Linux installation guide (NVIDIA#9485) Signed-off-by: Yiqing Yan <[email protected]> Co-authored-by: Yanchao Lu <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9075][doc] refine the slurm examples (NVIDIA#9548) Signed-off-by: Yan Chunwei <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9093][doc] update hyper links in overview (NVIDIA#9568) Signed-off-by: junq <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [TRTLLM-9092][doc] link to modelopt checkpoints in quick start guide (NVIDIA#9571) Signed-off-by: junq <[email protected]> Signed-off-by: Mike Iovine <[email protected]> Signed-off-by: Mike Iovine <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [None][fix] Fix triton moe load_weight (NVIDIA#9649) Signed-off-by: shuyix <[email protected]> [None][fix] fix a bug: deepseek_fp8_block_scales in TRTLLMGEN-MoE use 2D x_sf instead of 1D (NVIDIA#9658) Signed-off-by: xxi <[email protected]> [TRTLLM-9372][feat] Enable CuteDSL MoE with Large EP (NVIDIA#9592) Signed-off-by: Enwei Zhu <[email protected]> [TRTLLM-9522][chore] implement default `attach_multimodal_embeddings` (NVIDIA#9664) Signed-off-by: ixlmar <[email protected]> [TRTLLM-9660][feat] Convert cuteDSL GEMM to opt-in feature (NVIDIA#9682) Signed-off-by: Jonas Li <[email protected]> Co-authored-by: Kaiyu Xie <[email protected]> [None][fix] enable hmac in RPC (NVIDIA#9745) Signed-off-by: Superjomn <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [https://nvbugs/5703953][fix] Preserving ip:port for trtllm-serve before initializing llm (NVIDIA#9646) Signed-off-by: Junyi Xu <[email protected]> [None][infra] Waive failed cases for main branch on 12/07 (NVIDIA#9769) Signed-off-by: qqiao <[email protected]> [None][fix] Several minor fixes to CI setting (NVIDIA#9765) Signed-off-by: Yanchao Lu <[email protected]> [OMNIML-3036][doc] Re-branding TensorRT-Model-Optimizer as Nvidia Model-Optimizer (NVIDIA#9679) Signed-off-by: Chenjie Luo <[email protected]> [None][feat] Enable NCCL_SYMMETRIC as default fallback for AllReduce (NVIDIA#9314) Signed-off-by: Ludwig Schneider <[email protected]> [TRTLLM-9000][feat] Add multi-node Perf Tests into CI (NVIDIA#8800) Signed-off-by: Chenfei Zhang <[email protected]> [None][test] add ntp tolerance in time metrics verification (NVIDIA#9741) Signed-off-by: zhengd-nv <[email protected]> [TRTLLM-9603][feat] Enable ConfigurableMoE test in the CI (NVIDIA#9645) [https://nvbugs/5422621][test] Add GB 200 WIDEEP test case for RCCA 5422621 (NVIDIA#9506) Signed-off-by: FredricZ-2007 <[email protected]> [None][fix] Fix two tuning cache miss issues. (NVIDIA#9743) Signed-off-by: Yukun He <[email protected]> [None][infra] Check in most recent lock file from nightly pipeline Signed-off-by: TensorRT LLM <[email protected]> [TRTLLM-9706] [doc] Update wide EP documents (NVIDIA#9724) Signed-off-by: Kaiyu Xie <[email protected]> [https://nvbugs/5666804][test] only adding sampler config for limited models (NVIDIA#9512) Signed-off-by: Ruodi Lu <[email protected]> Co-authored-by: Ruodi Lu <[email protected]> Co-authored-by: yufeiwu-nv <[email protected]> Co-authored-by: Larry Xu <[email protected]> [None][infra] Waive failed cases for main on 12/08 (NVIDIA#9773) Signed-off-by: qqiao <[email protected]> [None][chore] Move the rocketkv e2e test to post-merge (NVIDIA#9768) Signed-off-by: Fanrong Li <[email protected]> [None][chore] Enable tvm_ffi for cute dsl nvfp4_gemm to reduce host overhead. (NVIDIA#9690) Signed-off-by: Mindy Li <[email protected]> [TRTLLM-9431][perf] Enable multistream for Linear Attention in Qwen3-… (NVIDIA#9696) Signed-off-by: nv-guomingz <[email protected]> [None][chore] Remove closed bugs (NVIDIA#9770) Signed-off-by: xinhe-nv <[email protected]> [None][infra] update mooncake in docker images (NVIDIA#9584) Signed-off-by: zhengd-nv <[email protected]> Signed-off-by: Zheng Duan <[email protected]> [None][test] Add Kimi k2 WIDEEP perf and accuracy cases (NVIDIA#9686) Signed-off-by: FredricZ-2007 <[email protected]> Signed-off-by: Kaiyu Xie <[email protected]> Co-authored-by: Kaiyu Xie <[email protected]> [https://nvbugs/5527655][test] Add test case for RCCA 5527655 (NVIDIA#9511) Signed-off-by: FredricZ-2007 <[email protected]> [http://nvbugs/5649010][fix] fix test_auto_scaling.py::test_worker_restart timeout (NVIDIA#9775) Signed-off-by: Lizhi Zhou <[email protected]> [None][fix] Switch AutoDeploy's default allreduce strategy to NCCL (NVIDIA#9666) Signed-off-by: Eran Geva <[email protected]> [TRTLLM-9506][fix] Fix AR for DeepSeek-R1 2 model path (NVIDIA#9661) Signed-off-by: qgai <[email protected]> ray + updatew works trtllm works in async env trtllm works in sync and async env ray + updatew works rebase to the updated verl server mode still cherry pick still cherry pick still cherry pick integrated http interface hang at RyExecutor create workers ray.remote clean code use tensorrt_llm.rlhf_utils Signed-off-by: Liwei Ma <[email protected]> placement, asyncllm, and basic tests Signed-off-by: Erin Ho <[email protected]> connect sleep and wakeup; Add support to pass None to update_weights Signed-off-by: Erin Ho <[email protected]> Batching ctx for IFB scheduler Signed-off-by: Yuan Tong <[email protected]> accuracy WAR for TP>1: always use AllReduceStrategy.NCCL, refactored Signed-off-by: Erin Ho <[email protected]> fix e2e integration Signed-off-by: Superjomn <[email protected]> update asyncllm, other nits Signed-off-by: Erin Ho <[email protected]> fix init setup Signed-off-by: Erin Ho <[email protected]> Fix TRTLLMSampler logprobs perf Signed-off-by: Yuan Tong <[email protected]> fix and cleanup Signed-off-by: Erin Ho <[email protected]> fix server Signed-off-by: Erin Ho <[email protected]> Revert "Batching ctx for IFB scheduler" This reverts commit b51aac0 Signed-off-by: Yuan Tong <[email protected]> update & address comments Signed-off-by: Erin Ho <[email protected]>
usberkeley
pushed a commit
to usberkeley/TensorRT-LLM
that referenced
this pull request
Dec 11, 2025
Signed-off-by: Tal Cherckez <[email protected]> Signed-off-by: tcherckez-nvidia <[email protected]> Co-authored-by: Neta Zmora <[email protected]>
codego7250
pushed a commit
to codego7250/TensorRT-LLM
that referenced
this pull request
Dec 11, 2025
Signed-off-by: Tal Cherckez <[email protected]> Signed-off-by: tcherckez-nvidia <[email protected]> Co-authored-by: Neta Zmora <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add handling for Lllama4 MoE where experts are calculated in BMM with swiglu
Perf is up by ~2.5x compared to baseline, there is still a gap of 35% to trtllm, probably due to comms.
Added tests
Summary by CodeRabbit
Release Notes
New Features
Tests
✏️ Tip: You can customize this high-level summary in your review settings.
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
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.
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.