kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-01 16:51:11 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
3,714 Commits 28 Branches 64 Tags 2 GiB
cd44f80abd
Commit Graph

248 Commits

Author SHA1 Message Date
sychen52
98a1bffb7c
[OMNIML-2336][feat] Add NVFP4 x FP8 (#6809) 
Signed-off-by: Shiyang Chen <shiychen@nvidia.com>
 2025-09-04 09:03:38 -07:00
Enwei Zhu
1745102e72
[TRTLLM-7027][feat] Fuse d2t to logitsBitmaskKernel and fix a race condition in one-model spec (#7481) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
Co-authored-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-09-04 23:30:14 +08:00
Yilin Fan
261ffacfa4 [https://nvbugs/5412562][feat] Allocate MoE workspace only when necessary (release/1.0 retargeted) (#6955) 
Signed-off-by: Yilin Fan <206948969+nv-yilinf@users.noreply.github.com>
Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
 2025-09-01 11:02:31 +08:00
Tian Zheng
e257cb3533
[None][feat] Support NVFP4 KV Cache (#6244) 
Signed-off-by: Tian Zheng <29906817+Tom-Zheng@users.noreply.github.com>
 2025-09-01 09:24:52 +08:00
Zongfei Jing
53163bf1df
[TRTLLM-6876][feat] Add low precision all2all for mnnvl (#7155) 
Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>
 2025-08-28 18:26:16 +08:00
Jin Li
028235404b
[TRTLLM-6633][feat] Padding for piecewise cudagraph (#6750) 
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-08-26 18:31:33 -04:00
Void
040f4c70d3
[None][perf] Accelerate global scale calculations for deepEP fp4 combine (#7126) 
Signed-off-by: Yilin Zhang <18275976+yilin-void@users.noreply.github.com>
 2025-08-27 00:13:13 +08:00
Bo Li
bf1b958f1a
[TRTLLM-7319][perf] Fuse slicing into MoE. (#6728) 
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
Signed-off-by: Sergey Klevtsov <sklevtsov@nvidia.com>
Co-authored-by: Sergey Klevtsov <sklevtsov@nvidia.com>
 2025-08-25 16:52:30 -04:00
dongxuy04
19a0ea363b
[TRTLLM-6743][feat] Optimize and refactor alltoall in WideEP (#6973) 
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
Signed-off-by: Fred Wei <20514172+WeiHaocheng@users.noreply.github.com>
Signed-off-by: Dongxu Yang <dongxuy@nvidia.com>
Co-authored-by: Fred Wei <20514172+WeiHaocheng@users.noreply.github.com>
 2025-08-24 08:15:29 -04:00
Daniel Stokes
f7c597ec40
[None][perf] Make finalize fusion part of the tactic selection logic (#6915) 
Signed-off-by: djns99 <40156487+djns99@users.noreply.github.com>
 2025-08-21 14:08:03 -07:00
Yuan Tong
90bfc8cc29
[https://nvbugs/5453827][fix] Fix RPATH of th_common shared library to find pip-installed NCCL (#6984) 
Signed-off-by: Yuan Tong <13075180+tongyuantongyu@users.noreply.github.com>
 2025-08-21 17:58:30 +08:00
ChristinaZ
c7269ea93a
[https://nvbugs/5392414] [fix] Add customized default routing method (#6818) 
Signed-off-by: Christina Zhang <83400082+ChristinaZ@users.noreply.github.com>
 2025-08-21 16:58:41 +08:00
zhhuang-nv
7e135d2ea7
[None][feat] Use Separate QKV Input Layout for Context MLA (#6538) 
Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>
 2025-08-19 22:04:48 +08:00
amitz-nv
a54c53652b
[TRTLLM-7263][fix] Prevent recreation of cublas handles in lora_grouped_gemm every call (#6968) 
Signed-off-by: Amit Zuker <203509407+amitz-nv@users.noreply.github.com>
 2025-08-19 15:39:56 +03:00
bhsueh_NV
85cbd0263b
[None][feat] Support Yarn on Qwen3 (#6785) 
Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>
 2025-08-17 07:21:29 +08:00
Yuening Li
1f8ae2b2db
[TRTLLM-5863][feat] Support MoE INT8 Weight-Only-Quantization in PyTorch Workflow (#6629) 
Signed-off-by: Yuening Li <62227368+yueningl@users.noreply.github.com>
 2025-08-15 17:15:49 -04:00
Liao Lanyu
f7c13a4aa7
[TRTLLM-6906][chore] Using pybind to bind functions in thop/attentionOp (#6745) 
Signed-off-by: Lanyu Liao <lancelly@users.noreply.github.com>
 2025-08-12 16:45:16 +08:00
Sergey Klevtsov
27fc35175e
[None][feat] CUTLASS MoE FC2+Finalize fusion (#3294) 
Signed-off-by: Sergey Klevtsov <sklevtsov@nvidia.com>
 2025-08-12 15:56:48 +08:00
NVJiangShao
2f2f5cc72c
[TRTLLM-6744][feat] Remove input_sf swizzle for module WideEPMoE (#6231) 
Signed-off-by: Jiang Shao <91270701+StudyingShao@users.noreply.github.com>
 2025-08-08 11:13:42 +08:00
Iman Tabrizian
82276167e6
[None][feat] Add NCCL Symmetric Integration for All Reduce (#4500) 
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
 2025-08-07 17:28:14 -07:00
peaceh-nv
8ec3b1de10
[None][feat] : Add FP8 context MLA support for SM120 (#6059) 
Signed-off-by: peaceh <103117813+peaceh-nv@users.noreply.github.com>
 2025-08-07 16:16:34 +08:00
hlu1
8207d5fd39
[None] [feat] Add model gpt-oss (#6645) 
Signed-off-by: Hao Lu <14827759+hlu1@users.noreply.github.com>
 2025-08-07 03:04:18 -04:00
Haohang Huang
c9eebcb454
[TRTLLM-6674][feat] (Breaking Change) Hopper SWA non-cyclic kernels + KV reuse + Spec Dec (#6379) 
Signed-off-by: Haohang Huang <31998628+symphonylyh@users.noreply.github.com>
Signed-off-by: symphonylyh <31998628+symphonylyh@users.noreply.github.com>
 2025-08-05 07:47:41 +00:00
Yuan Tong
a2f271c8e0
[TRTLLM-4406][feat] LLM sleep & wakeup Part 1: virtual device memory (#5034) 
Signed-off-by: Yuan Tong <13075180+tongyuantongyu@users.noreply.github.com>
 2025-08-04 13:51:01 +08:00
Jhao-Ting Chen
6edaa23c1c
[None][feat] Multi-block mode for Hopper spec dec XQA kernel (#4416) 
Signed-off-by: Jhao-Ting Chen <jhaotingc@nvidia.com>
 2025-08-03 14:31:33 -07:00
Yang Li
ac23f4a80d
[TRTLLM-4279] fix: Add a protection test for checking trtllm custom ops (#6515) 
Signed-off-by: Yang Li <56944310+yali-arch@users.noreply.github.com>
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
 2025-08-01 15:59:09 +08:00
liji-nv
e07fff4f78
[https://nvbugs/5340941] - fix: Correct custom ops used by Qwen3 Moe … (#6285) 
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-07-25 14:49:45 +08:00
Zhenhua Wang
62298bc473
perf: customize cublastLt algo for Llamba 3.3 70B TP4 (#6315) 
Signed-off-by: Zhenhua Wang <zhenhuaw@nvidia.com>
 2025-07-24 23:01:15 +08:00
Perkz Zheng
2193ad3aac
[https://nvbugs/5387771] fix deadlocks due to insufficient numSemaphores (#6262) 
Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>
 2025-07-23 11:20:55 +08:00
Yuening Li
e8c068b4b1
[TRTLLM-5863][feat] Support Weight-Only-Quantization in PyTorch Workflow (#5850) 
Signed-off-by: Yuening Li <62227368+yueningl@users.noreply.github.com>
Co-authored-by: Yuening Li <62227368+yueningl@users.noreply.github.com>
 2025-07-21 15:17:35 +08:00
danielafrimi
5300a99bd8
W4A8 GEMM (#6005) 
Signed-off-by: Daniel Afrimi <danielafrimi8@gmail.com>
 2025-07-20 17:34:57 +03:00
bhsueh_NV
2e14c8f443
[Fix][Chore][Qwen3] fix bug of using fp4 on sm120 (#6065) 
Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>
 2025-07-20 10:25:25 +08:00
Enwei Zhu
21efb50068
[TRTLLM-6406] feat: Enable guided decoding with overlap scheduler (#6000) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-07-17 17:46:10 +08:00
Jinyang Yuan
e761231c0b
[fix] Move NCCL group in all-gather and reduce-scatter OPs outside the outer loop (#6053) 
Signed-off-by: Jinyang Yuan <154768711+jinyangyuan-nvidia@users.noreply.github.com>
 2025-07-16 00:25:32 +09:00
Yi Zhang
9cc4e5d50e [nvbugs/5336321][fix] Enable attention dp = False test case, Fix TRTLLM Gen Moe workspace allocation (#5463) 
Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
Signed-off-by: yizhan <187001205+yizhang-nv@users.noreply.github.com>
 2025-07-14 17:17:30 +08:00
Dom Brown
afaa388bee [TRTLLM-6100] fix: Nvbug 5356427: autotuned TRTLLM Gen fp8 block scale MoE illegal memory access (#5676) 
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
 2025-07-14 17:17:30 +08:00
Enwei Zhu
bc1d4fb5da
[NvBug 5378370] fix: Fix alltoall for llama4 (apply_router_weight_on_input=True) (#5902) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-07-12 15:50:31 +09:00
Jinyang Yuan
8b9a030a5c
[fix] Fix MoE workspace info by storing Torch tensor itself instead of data_ptr (#5900) 
Signed-off-by: Jinyang Yuan <154768711+jinyangyuan-nvidia@users.noreply.github.com>
 2025-07-10 20:07:32 +09:00
Anthony Chang
7d21b55b5a
[feat] Add TRTLLM MoE nvfp4 cubins for mid-high concurrency; attention_dp for TRTLLM MoE (#5723) 
Signed-off-by: Anthony Chang <27950904+rosenrodt@users.noreply.github.com>
 2025-07-10 14:06:50 +08:00
QI JUN
e289a98d5a
avoid nesting NCCL group in allgather and reduce scatter OPs (#5866) 
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
 2025-07-10 12:32:59 +09:00
DylanChen-NV
74dca0aa7b
[NVBUG-5304516/5319741]Qwen2.5VL FP8 support (#5029) 
Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
 2025-07-09 23:16:42 +08:00
Dom Brown
3e3b1769ad
[TRTLLM-5881] feat: Integrate TRT-LLM Gen FP4 block scale MoE with Pytorch workflow kernel autotuner (#5764) 
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
 2025-07-09 08:21:58 +01:00
JieXin Liang
664bf95892
[fix] improve fp4_block_scale_moe_runner type check (#5681) 
Signed-off-by: JieXin Liang <Alcanderian@users.noreply.github.com>
Co-authored-by: ChristinaZ <83400082+ChristinaZ@users.noreply.github.com>
 2025-07-08 14:32:14 +09:00
DylanChen-NV
5ca2b9bb15
[TRTLLM-5812][feat] support FP8 row-wise dense GEMM in torch flow (#5615) 
Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
 2025-07-07 18:04:57 +08:00
Daniel Stokes
ec6c7dff1a
feat: Add support for MXFP8xMXFP4 in pytorch (#5535) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-07-06 15:32:06 -07:00
Jhao-Ting Chen
77082cde38
[https://nvbugspro.nvidia.com/bug/5329655] [feat] Pytorch path add spec dec param to attention op (#5146) 
Signed-off-by: Jhao-Ting Chen <jhaotingc@nvidia.com>
 2025-07-02 04:54:43 -04:00
Void
7992869798
perf: better heuristic for allreduce (#5432) 
Signed-off-by: Yilin Zhang <18275976+yilin-void@users.noreply.github.com>
 2025-07-01 22:56:06 -04:00
liji-nv
c345f5876c
[feat] Support torch compile for attention dp (#5086) 
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-07-01 13:48:52 -04:00
danielafrimi
7a617ad1fe
feat: W4A16 GEMM (#4232) 
Signed-off-by: Daniel Afrimi <danielafrimi8@gmail.com>
 2025-07-01 10:36:05 +03:00
Li Min
16fc99391f
refactor: [TRTLLM-6150] Refactor moe permute and finalize op by removing duplicated code (#5557) 
Signed-off-by: Mindy Li <11663212+limin2021@users.noreply.github.com>
 2025-06-30 08:48:04 -07:00
WeiHaocheng
42a9385d02
[TRTLLM-5331] perf: Replace allgaher with AllToAllPrepare (#5570) 
Signed-off-by: Fred Wei <20514172+WeiHaocheng@users.noreply.github.com>
 2025-06-30 13:06:09 +08:00
Enwei Zhu
b4dab23e7b
[TRTLLM-5965] perf: Optimize MoE sort kernels for large-scale EP (#5435) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-06-30 01:02:07 +08:00
Li Min
6021a439ab
Make moe permute and final as custom op (#5412) 
Signed-off-by: Mindy Li <11663212+limin2021@users.noreply.github.com>
 2025-06-27 15:48:33 -07:00
Daniel Stokes
5773cfdcf2
feat: Add support for per expert activation scaling factors (#5013) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-06-28 09:10:35 +12:00
ChristinaZ
a608b00d38
Fix mPtrExpertCounts allocation in MoE TRT-LLM backend (nvfp4) (#5519) 
Signed-off-by: Christina Zhang <83400082+ChristinaZ@users.noreply.github.com>
 2025-06-27 20:17:40 +08:00
Daniel Stokes
83a1f60556
feat: Expose bias and FP8_MXFP4 MOE CUTLASS backend features to pytorch (#5410) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-06-27 12:29:34 +08:00
jmydurant
8836990bde
[TRTLLM-3602][feat] support nvfp4 model and fp8 kv cache for MLA chunked prefill (Blackwell) (#5475) 
Signed-off-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com>
 2025-06-26 22:18:08 +08:00
Bo Li
1bab9000a6
perf: Optimize swizzle_sf, unswizzle_sf, reswizzle_sf (#5318) 
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
 2025-06-26 14:03:56 +08:00
dongxuy04
490d2e5819
feat: large-scale EP(part 8: Online EP load balancer integration for PCIe fp8) (#5226) 
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
 2025-06-25 22:25:13 -07:00
Daniel Stokes
942841417e
opensource: Opensource MOE MXFP8-MXFP4 implementation (#5222) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-06-26 12:18:19 +08:00
jmydurant
578dbc8d9a
feat: chunked prefill for MLA (Blackwell) (#4651) 
Signed-off-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com>
 2025-06-26 09:01:00 +08:00
dongxuy04
4f0f17ac8a
feat: Misc Opt for large scale EP (#5374) 
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
 2025-06-20 13:11:31 +08:00
Fanrong Li
5d4ab47d5b
fix: refactor and fix mtp vanilla (#4762) 
Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
 2025-06-20 05:23:39 +08:00
Fanrong Li
c7af650d5a
Fix: fix the deterministic issue in the MTP Eagle path (#5285) 
Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
 2025-06-19 18:08:40 +08:00
Bo Li
d76bda7f2c
chore: Refine printed info of CHECK_TYPE. (#5295) 
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
 2025-06-18 15:35:41 +08:00
Yukun He
6711ad9cf3
[TRTLLM-5589] feat: Minor optimizations for tunable FP8 batched GEMM op. (#5139) 
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
 2025-06-18 14:33:46 +08:00
Dom Brown
44fb3c1673
[TRTLLM-5770] feat: Integrate TRT-LLM Gen FP8 block scale MoE with Pytorch workflow kernel autotuner (#5207) 
- Adds a new Python custom op (fp8_block_scale_moe_runner) and a FP8BlockScaleMoERunner class for autotuning.
- Updates C++ MoE and batched GEMM kernels to accept a configIndex for workspace sizing and execution.
- Extends the unit test to run both autotuned and non-autotuned code paths.

Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
 2025-06-17 21:01:56 +08:00
liji-nv
13eef642e6
[feat] Piecewise cuda graph support for MLA (#4467) 
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-06-17 18:58:38 +08:00
Enwei Zhu
4b82b8b4c7
[TRTLLM-5330] perf: Optimize MoE supplementary kernels for large-scale EP (#5215) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-06-17 15:23:24 +08:00
Tracin
ef3fdc8051
feat: Add w4a8_mxfp4_fp8 quantization recipe. (#4867) 
Signed-off-by: Tracin <10434017+Tracin@users.noreply.github.com>
 2025-06-16 11:30:57 +08:00
yunruis
b99c5ce8c1
Feat/ds r1 min latency opt round3, add router gemm, fused a gemm, PDL (#4560) 
Signed-off-by: yunruis <yunruis@nvidia.com>
Signed-off-by: kduan <176893526+Kefeng-Duan@users.noreply.github.com>
Signed-off-by: Kefeng-Duan <176893526+Kefeng-Duan@users.noreply.github.com>
Co-authored-by: kduan <176893526+Kefeng-Duan@users.noreply.github.com>
 2025-06-14 17:36:22 +08:00
yunruis
e5be3a95b3
fix: fix license bug (#5200) 
Signed-off-by: yunruis <205571022+yunruis@users.noreply.github.com>
 2025-06-13 18:58:15 +08:00
yunruis
30c5b4183a
refactoring: port customized kernels with public cutlass version (#5027) 
Signed-off-by: yunruis 

Merge this to unblock others since the full CI has been run through
 2025-06-13 16:19:31 +08:00
zhhuang-nv
a891013e3c
[feat] Optimize KV Cache Reuse for MLA (#4869) 
Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>
 2025-06-13 11:03:05 +08:00
Matthias Jouanneaux
a0b6c635b1
[feat] trtllmGen MoE routing: added support for top groups and top K bounds (#4063) 
Signed-off-by: Matthias Jouanneaux <mjoux@nvidia.com>
Co-authored-by: hlu1 <14827759+hlu1@users.noreply.github.com>
Co-authored-by: Nikita Korobov <14355239+nekorobov@users.noreply.github.com>
 2025-06-13 06:00:02 +08:00
HuiGao-NV
43192379af
Use backend to replace macro to control enablement of MNNVL all reduce (#4635) 
Signed-off-by: Hui Gao <huig@nvidia.com>
 2025-06-12 11:22:49 +08:00
Zongfei Jing
6d1f2d0fd7
[TRTLLM-3927] [feat] Finalize + Allreduce + add + rmsnorm fusion (#4756) 
Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>
 2025-06-10 19:55:16 +08:00
Dom Brown
9c012d5bf8
[TRTLLM-5589] feat: Integrate TRT-LLM Gen FP8 Batched GEMM with Pytorch workflow kernel autotuner (#4872) 
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
 2025-06-09 11:02:48 +01:00
ChristinaZ
f45aff2b7d
Add customized renormalized moe routing kernel for moe cutlass backend (#4955) 
Signed-off-by: Christina Zhang <83400082+ChristinaZ@users.noreply.github.com>
 2025-06-09 17:38:50 +08:00
Daniel Stokes
3a4851b7c3
feat: Add Mixture of Experts FP8xMXFP4 support (#4750) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-06-09 13:25:04 +08:00
dongxuy04
1e369658f1
feat: large-scale EP(part 6: Online EP load balancer integration for GB200 nvfp4) (#4818) 
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
Signed-off-by: ShiXiaowei02 <39303645+Shixiaowei02@users.noreply.github.com>
Co-authored-by: ShiXiaowei02 <39303645+Shixiaowei02@users.noreply.github.com>
 2025-06-08 10:25:18 +08:00
Shiyu Li
b0d287c9b7
[TRTLLM-4647][fix] Fix the no fusion allreduce hanging (#4594) 
Signed-off-by: Shiyu Li <shili@nvidia.com>
 2025-06-04 18:26:13 -07:00
Nikita Korobov
8043d7a03c
feat: update DeepSeek FP8 TRT-LLM Gen cubins (#4643) 
Signed-off-by: Nikita Korobov <nkorobov@nvidia.com>
 2025-06-03 14:07:54 -07:00
Tian Zheng
9832787050
[feat] Enable NVFP4 output for TRTLLM attention kernels (#4737) 
Signed-off-by: Tian Zheng <29906817+Tom-Zheng@users.noreply.github.com>
 2025-06-03 10:00:17 +08:00
tomeras91
bf9cd11fd4
[TRTLLM-4783][feat] Mamba2 kernel updates for Nemotron-H (#4494) 
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
 2025-06-01 13:56:44 +03:00
Enwei Zhu
25dde49c28
fix: EP load balancer with MTP layer and route offset by EP rank (#4767) 
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-06-01 00:07:44 +08:00
Jinyang Yuan
5339d367ce
[perf] Reduce the workspace size of FP4 activation scales for MoE (#4303) 
Signed-off-by: Jinyang Yuan <154768711+jinyangyuan-nvidia@users.noreply.github.com>
 2025-05-30 09:03:52 +08:00
Yilin Fan
31bb650298
Cherry pick feat/llama4 to main (#4739) 
Signed-off-by: Chenfei Zhang <chenfeiz@nvidia.com>
Signed-off-by: Yilin Fan <206948969+nv-yilinf@users.noreply.github.com>
Co-authored-by: Chenfei Zhang <chenfeiz@nvidia.com>
 2025-05-30 05:28:40 +08:00
ixlmar
fbe4db207d
feat: forward exceptions to Python and catch OOMs (#4497) 
Signed-off-by: ixlmar <206748156+ixlmar@users.noreply.github.com>
 2025-05-28 11:58:10 +02:00
zhhuang-nv
8452775db8
[TRTLLM-5070][feat] Support FP8 KV Cache Reuse for MLA (#4535) 
* optimize kv cache reuse workflow for MLA

write kv cache first and only call up-projection GEMM once
relax contiguous requirements of k/v for setting paged kv cache
return two contiguous tensors when loading MLA KV Cache

Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>

* support fp8 kv cache for MLA kv cache reuse

Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>

* resolve comments

Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>

---------

Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com>
 2025-05-23 19:47:50 +08:00
Anthony Chang
bbea2647b1
Qwen3 supports TRTLLM FP4 MoE backend (#4530) 
* MoE TRTLLM backend for Qwen3

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* add extra moe_backend to test

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* address comments

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* conditionally compile kernels on newer archs

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* missing positional arg

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* Update the routing kernels

Signed-off-by: Christina Zhang <christinaz@nvidia.com>

* Revise usage of TLLM_LOG_ERROR

Signed-off-by: Christina Zhang <christinaz@nvidia.com>

* Add unit test for Qwen3 moe (trtllm_gen backend)

Signed-off-by: Christina Zhang <christinaz@nvidia.com>

* improve weight processing speed of moe_backend=TRTLLM; roughly 2x

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* tidy and minor fix

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

* temporarily disable accuracy test that has known issue

Signed-off-by: Anthony Chang <anchengc@nvidia.com>

---------

Signed-off-by: Anthony Chang <anchengc@nvidia.com>
Signed-off-by: Christina Zhang <christinaz@nvidia.com>
Co-authored-by: Christina Zhang <christinaz@nvidia.com>
 2025-05-23 18:31:08 +08:00
Bo Li
9ae705af1b
perf: Add fused q_norm/k_norm/RoPE for Qwen3. (#4482) 
* Add Julien's origina kernel.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Get rid of UpdateKVCache functionality.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add kernels.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add torch OP.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Update cmake.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Torch OP must use double as argument dtype.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add unittest.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add unittest.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Fix misaligned access when head_dim=64.
In this case, numElemsPerThread=2, numVecPerThread=0. But the store code incorrectly perform vectorized store, some threads (e.g., lane1) issue store to address that is not aligned to 64 bit.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Remove unroll (compiler can do that).
Cleanup code.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add switch for interleave.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Refactor vectorized load/store.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Implement is_neox. Result not correct yet.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Fix is_neox=True.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

* Add q_weight and k_weight.

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>

---------

Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
 2025-05-23 15:31:04 +08:00
CarstyYou
ef280e687e
[feat] support fp8 blockscale gemm on sm89 (#4481) 
* [feat] integrate ada blockwise gemm

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [fix] align scale M

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [feat] swizzle mma output

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [test] add ut for sm89

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [delete] remove useless comments

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [chore] codestyle

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [fix] fix review comments

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [chore] fix license

Signed-off-by: CarstyYou <xiy@nvidia.com>

* [chore] fix license

Signed-off-by: CarstyYou <xiy@nvidia.com>

---------

Signed-off-by: CarstyYou <xiy@nvidia.com>
Co-authored-by: bhsueh_NV <11360707+byshiue@users.noreply.github.com>
 2025-05-23 10:39:10 +08:00
Mike Iovine
9c0de251db
[feat] Integrate Hopper chunked attention kernels (#4330) 
* Integrate chunked attention kernels

Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>

* Fix cache key

Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>

* Fix lint

Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>

---------

Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
 2025-05-22 17:10:57 -04:00
Zongfei Jing
dbaddb3a29
Adding two-shot allreduce kernel and mnnvl multicasting buffer (#4216) 
* Adding two-shot allreduce kernel and mnnvl multicasting buffergit gffe

Signed-off-by: Shiyu Li <shili@nvidia.com>

Adding comments

Signed-off-by: Shiyu Li <shili@nvidia.com>

Add unittest of the twoshot kernel.

Signed-off-by: Shiyu Li <shili@nvidia.com>

Update dispatch logic

Signed-off-by: Shiyu Li <shili@nvidia.com>

Use cpu barrier instead of GPU at init

Signed-off-by: Shiyu Li <shili@nvidia.com>

Merge dispatch logic fix

Signed-off-by: Shiyu Li <shili@nvidia.com>

Update the kernel to use GPU-managed buffer

Signed-off-by: Shiyu Li <shili@nvidia.com>

* Refine

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Clean code

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Fix compile error

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Fix issue

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Clean up

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Simplify AllReduce interface

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Rename

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Fix warning

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Tidy code

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Rename

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Fix compile error

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Refine

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Skip ut for no_fusion

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Refine

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

---------

Signed-off-by: Shiyu Li <shili@nvidia.com>
Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>
Co-authored-by: Shiyu Li <shili@nvidia.com>
 2025-05-22 03:42:36 +08:00
djns99
a030a898d1
perf: Fuse gemm setup function for SM90/SM100 MOE plugin path (#4146) 
Signed-off-by: Daniel Stokes <40156487+djns99@users.noreply.github.com>
 2025-05-21 10:00:36 +08:00
dongxuy04
21aff2e313
feat: large-scale EP(part 2: MoE Load Balancer - core utilities) (#4384) 
* first commit of cpp moe loadbalance code

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>

* add python bindings for moe load balance

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>

* add python wrapper, ut and bug fixes

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>

* add binding for layerId and update binding test

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>

* add host tensor sharing and ut

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>

---------

Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
 2025-05-20 17:53:48 +08:00
kanghui0204
6f3922f318
feat: Low Precision Allreduce for PCIe based GPU (#4344) 
This PR adds a customized allreduce to TensorRT-LLM. The new allreduce is used for communication on PCIe-based GPUs via low-precision quantization, which can accelerate the PCIe allreduce process.

Signed-off-by: Hui Kang <hkang@nvidia.com>
Co-authored-by: Hui Kang <hkang@nvidia.com>
 2025-05-20 06:53:46 +08:00
liji-nv
58e405624a
[https://nvbugs/5123103][fix] Fix torch compile for DeepSeekV3 (#3952) 
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
 2025-05-19 22:12:25 +08:00
Jinyang Yuan
b618e1f55b
perf: Eliminate the need for attention DP padding when possible (#3439) 
Signed-off-by: Jinyang Yuan <154768711+jinyangyuan-nvidia@users.noreply.github.com>
Co-authored-by: raccoonliukai <raccoonliu@tencent.com>
 2025-05-17 13:30:55 +08:00