biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity
14,985 Commits 2 Branches 140 Tags
738d0a281fab2e151a67b370c26b4e4360362f8f
Commit Graph

71 Commits

Author SHA1 Message Date
Wentao Ye
0ef7f79054 [Perf] Add tuned triton moe config for Qwen3.5 H200, 9.9% E2E throughput improvement (#37340) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-18 14:18:34 -04:00
Yan Ma
894843eb25 replace with torch.cuda.device with with torch.accelerator.device_index (#36144) 
Signed-off-by: Yan Ma <yan.ma@intel.com>
 2026-03-11 23:12:57 -07:00
Kunshang Ji
66a2209645 [Hardware] Replace torch.cuda.synchronize() api with torch.accelerator.synchronize (#36085) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-03-05 10:36:39 +00:00
Kunshang Ji
16d2ad1d38 [Hardware] Replace torch.cuda.empty_cache with torch.accelerator.empty_cache (#30681) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-03-04 09:49:47 +00:00
Robert Shaw
97995f6376 [MoE Refactor] Create MK for TRTLLM Kernels (#32564) 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Signed-off-by: Robert Shaw <rshaw@neuralmagic.com>
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <rshaw@neuralmagic.com>
 2026-03-03 10:39:50 -08:00
Matthias Gehre
934acddef9 [Perf] fused_moe: add int4_w4a16 benchmark support and tuning config (#34130) 
Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
 2026-02-13 00:14:27 -08:00
Michael Goin
ff1f83b056 [Refactor] Replace activation: str with MoEActivation enum (#33843) 
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
 2026-02-11 17:29:32 -08:00
Matthias Gehre
7a048ee65f [Bugfix] Fix benchmark_moe.py inplace assertion with torch >= 2.9 (#34149) 
Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
 2026-02-11 03:58:56 +00:00
Jee Jee Li
978a37c823 [Model] GLM adaptation (#34124) 2026-02-09 17:32:52 +08:00
Dimitrios Bariamis
f0bca83ee4 Add support for Mistral Large 3 inference with Flashinfer MoE (#33174) 
Signed-off-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>
Co-authored-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
 2026-01-30 22:48:27 -08:00
danisereb
f999539869 Add missing import of fused_topk to benchmark_moe (#32784) 
Signed-off-by: Daniel Serebrenik <daserebrenik@nvidia.com>
 2026-01-21 18:30:10 +00:00
Robert Shaw
42135d6898 [MoE Refactor] Oracle Select FP8+NVFP4 Kernels In Priority (#32414) 2026-01-21 08:22:33 -05:00
Yuxuan Zhang
71832ba71e [GLM-4.7] GLM Model support for GLM-Lite (#31386) 
Signed-off-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Signed-off-by: Yuxuan Zhang <2448370773@qq.com>
 2026-01-19 01:18:38 -08:00
Robert Shaw
5dcd7ef1f2 [MoE Refactor][15/N] Apply Refactor to Fp8 (#31415) 2026-01-07 19:42:33 -05:00
Cyrus Leung
db318326a5 [Misc] Use deprecated for seed_everything (#31780) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-06 11:29:55 +00:00
Alfred
a0e9ee83c7 [Benchmark] Fix OOM during MoE kernel tuning for large models (#31604) 
Signed-off-by: Alfred <massif0601@gmail.com>
 2026-01-02 22:24:51 +00:00
Kunshang Ji
2a2d5d2780 Replace torch.cuda.Event with torch.Event for better hardware compatibility (#26985) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2025-11-18 11:34:36 -08:00
Xiake Sun
03fa4d3fb3 [Hardware][AMD][Model] Add Triton MoE tuning support and optimized configs for Qwen3 omni for MI308X (#28373) 
Signed-off-by: Xiake Sun <xiake.sun@amd.com>
Signed-off-by: Xiake Sun <xisun@amd.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2025-11-10 04:53:40 +00:00
tomeras91
e4ee658672 [Model] add optimal triton fused moe configs for NemotronH MoE (#27967) 
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
 2025-11-04 12:59:43 +00:00
Yeshwanth N
71b1c8b667 [Chore]:Extract math and argparse utilities to separate modules (#27188) 
Signed-off-by: Yeshwanth Surya <yeshsurya@gmail.com>
Signed-off-by: Yeshwanth N <yeshsurya@gmail.com>
Signed-off-by: yeshsurya <yeshsurya@gmail.com>
 2025-10-26 04:03:32 -07:00
wangxiyuan
8f4b313c37 [Misc] rename torch_dtype to dtype (#26695) 
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
 2025-10-15 12:11:48 +00:00
Lukas Geiger
338b1bf04f [Benchmarks] Add support for Qwen 3 VL MoE tuning (#26419) 
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
 2025-10-08 14:01:08 +00:00
Jee Jee Li
67f3fb0844 [Bench] Add DeepSeekV32 to MoE benchmark (#25962) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-09-30 14:13:48 -07:00
bnellnm
5963b98b46 [Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537) 
Signed-off-by: Bill Nell <bnell@redhat.com>
 2025-09-17 17:43:31 -06:00
Jee Jee Li
04ad0dc275 [benchmark] Add triton version in the moe tuned config (#24769) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-09-16 14:10:54 +08:00
Jee Jee Li
d11ec124a0 [Bench] Add qwen-next in benchmark_moe.py (#24661) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-09-11 21:29:43 +08:00
Jee Jee Li
62f66be1f7 [Bugfix] Fix Qwen3-coder moe tuned config (#24072) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-09-07 05:19:46 +00:00
YUQI.CHENG
66548f6603 [Bugfix] Fix benchmark_moe.py for blockwise fp8. (#23823) 
Signed-off-by: crischeng <420985011@qq.com>
Co-authored-by: cris <grace@guisenbindeMacBook-Pro.local>
 2025-08-28 21:44:09 +08:00
Jee Jee Li
4d9c61993a [Bugfix] Fix benchmark_moe.py (#23177) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-08-19 13:39:40 +00:00
Jee Jee Li
6d3da472bc [Misc] Add --save-dir option to benchmark_moe (#23020) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-08-16 07:26:10 +00:00
Jee Jee Li
384a052971 [Misc] benchmark_moe supports expert parallel (#22251) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-08-11 00:13:27 -07:00
Jee Jee Li
8d705996df [Misc] Minor enhancement of benchmark_moe (#22068) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-08-02 01:35:30 +08:00
Yuxuan Zhang
10eb24cc91 GLM-4 Update (#20736) 
Signed-off-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: Lu Fang <fanglu@fb.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Lu Fang <fanglu@fb.com>
 2025-07-19 22:40:31 +00:00
Asher
5a7fb3ab9e [Model] Add ToolParser and MoE Config for Hunyuan A13B (#20820) 
Signed-off-by: Asher Zhang <asherszhang@tencent.com>
 2025-07-17 09:10:09 +00:00
Wentao Ye
e2de455c34 [Feature] Integrate SM100 DeepGEMM support (#20087) 2025-07-10 20:18:05 -07:00
Brayden Zhong
cede942b87 [Benchmark] Add support for multiple batch size benchmark through CLI in benchmark_moe.py (#20516) 
Signed-off-by: Brayden Zhong <b8zhong@uwaterloo.ca>
 2025-07-06 09:20:11 +00:00
Wentao Ye
a6c4b87fbc Revert "[Feature] Integrate new deepgemm (#19820)" (#20049) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2025-06-24 19:45:22 -07:00
Wentao Ye
c6e3bba8e6 [Feature] Integrate new deepgemm (#19820) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2025-06-24 12:51:56 -07:00
Tianyu Guo
4589b94032 [Bugfix] Fix benchmark_moe.py (#19016) 
Signed-off-by: Tianyu Guo <guoty9@mail2.sysu.edu.cn>
 2025-06-09 18:04:36 -07:00
Simon Mo
02f0c7b220 [Misc] Add SPDX-FileCopyrightText (#19100) 
Signed-off-by: simon-mo <simon.mo@hey.com>
 2025-06-03 11:20:17 -07:00
Harry Mellor
009d9e7590 Convert benchmarks to ruff format (#18068) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2025-05-13 13:43:29 +00:00
xsank
0a9bbaa104 [Misc] support model prefix & add deepseek vl2 tiny fused moe config (#17763) 
Signed-off-by: 唯勤 <xsank.mz@alibaba-inc.com>
Co-authored-by: 唯勤 <xsank.mz@alibaba-inc.com>
 2025-05-08 07:50:22 +00:00
Mengqing Cao
f9bc5a0693 [Bugfix] Fix triton import with local TritonPlaceholder (#17446) 
Signed-off-by: Mengqing Cao <cmq0113@163.com>
 2025-05-06 17:53:09 +08:00
Xiaodong Wang
9352cdb56d [Hardware][AMD] Improve OAM device ID + llama4 Maverick MOE tuning (#16263) 
Signed-off-by: Lu Fang <lufang@fb.com>
Co-authored-by: Lu Fang <lufang@fb.com>
 2025-05-02 19:44:19 +00:00
Caleb_Du
3e887d2e0c permute/unpermute kernel for moe optimization (#14568) 
Signed-off-by: Caleb_Du <Caleb_Du@zju.edu.cn>
 2025-05-02 11:31:55 -07:00
Michael Goin
8fc88d63f1 [Model] Add tuned triton fused_moe configs for Qwen3Moe (#17328) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2025-04-28 15:20:24 -07:00
Harry Mellor
423e9f1cbe Use Transformers helper get_text_config() instead of checking for text_config (#17105) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2025-04-25 08:47:35 -07:00
Lu Fang
55dcce91df Upstream Llama4 Support to Main (#16113) 
Signed-off-by: Aston Zhang <22279212+astonzhang@users.noreply.github.com>
Signed-off-by: Chris Thi <chris.c.thi@gmail.com>
Signed-off-by: drisspg <drisspguessous@gmail.com>
Signed-off-by: Jon Swenson <jmswen@gmail.com>
Signed-off-by: Keyun Tong <tongkeyun@gmail.com>
Signed-off-by: Lu Fang <fanglu@meta.com>
Signed-off-by: Xiaodong Wang <xdwang@meta.com>
Signed-off-by: Yang Chen <yangche@fb.com>
Signed-off-by: Ye (Charlotte) Qi <yeq@meta.com>
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
Signed-off-by: Zijing Liu <liuzijing2014@gmail.com>
Signed-off-by: Lu Fang <lufang@fb.com>
Signed-off-by: Lu Fang <fanglu@fb.com>
Signed-off-by: Lucia Fang <fanglu@fb.com>
Signed-off-by: Roger Wang <ywang@roblox.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Lu Fang <fanglu@fb.com>
Co-authored-by: Roger Wang <ywang@roblox.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2025-04-07 08:06:27 -07:00
bnellnm
e59ca942f5 Add option to use DeepGemm contiguous grouped gemm kernel for fused MoE operations. (#13932) 
Signed-off-by: Bill Nell <bnell@redhat.com>
 2025-04-01 12:07:43 -04:00
Jee Jee Li
a73122de96 [Bugfix] fix benchmark moe (#14653) 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
 2025-03-13 16:12:42 +08:00