biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity
15,285 Commits 2 Branches 140 Tags
db01535e2bd5641e8cd7f88e5febf1d2a26bfbce
Commit Graph

2088 Commits

Author SHA1 Message Date
Zhewen Li
be1a85b7a2 Revert "[MoE Kernel] Flashinfer nvfp4 cutedsl moe kernel integration" (#38050) (#38169) 
Co-authored-by: Zhewen Li <zhewenli@inferact.ai>
 2026-03-26 07:59:09 -07:00
Vadim Gimpelson
52069012fe [Bugfix] Fix DeepGemm E8M0 accuracy degradation for Qwen3.5 FP8 on Blackwell (#38083) 
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
 2026-03-26 01:21:47 -07:00
BadrBasowid
e6bf9f15ec [Bugfix][CI] Fix Marlin FP8 Linear Kernel for Compressed Tensors Format (#38092) 
Signed-off-by: BadrBasowid <Badr.Basowid@gmail.com>
Signed-off-by: BadrBasowid <61441185+BadrBasowid@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2026-03-25 21:11:43 -07:00
Jacob Platin
d7d51a7ee5 [Bugfix] Fix Qwen3.5-FP8 Weight Loading Error on TPU (#37348) 
Signed-off-by: Jacob Platin <jacobplatin@google.com>
 2026-03-26 00:46:01 +00:00
Andreas Karatzas
7d6917bef5 [ROCm] Fix MoE kernel test failures on gfx950 (#37833) 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Signed-off-by: Matthew Wong <Matthew.Wong2@amd.com>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: Matthew Wong <Matthew.Wong2@amd.com>
 2026-03-25 13:46:40 -05:00
Yongye Zhu
678b3c99e8 [MoE Kernel] Flashinfer nvfp4 cutedsl moe kernel integration (#38050) 2026-03-25 10:16:40 -07:00
Kunshang Ji
14771f7150 [XPU] support MLA model on Intel GPU (#37143) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-03-25 17:43:42 +08:00
Chauncey
09c3dc9186 [Revert] Remove CUDA torch fallbacks for fp8_mqa_logits/fp8_paged_mqa_logits_torch function (#37968) 
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
 2026-03-25 06:19:37 +00:00
Andreas Karatzas
679c6a3ecc [Bugfix][ROCm][MoE] Fix mxfp4 oracle regressions from #37128 (#37787) 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
 2026-03-25 08:17:33 +08:00
Harry Mellor
b3601da6e7 [Mypy] Fix mypy for vllm/model_executor (except vllm/model_executor/layers) (#37904) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-03-24 17:14:01 +00:00
Li, Jiang
352b90c4a4 [Bugfix] Add replacement of _compute_slot_mapping_kernel on CPU (#37987) 
Signed-off-by: jiang1.li <jiang1.li@intel.com>
 2026-03-24 07:00:20 -07:00
Wentao Ye
c59a132f96 [V0 Deprecation] Refactor kv cache from list to element (#37487) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-23 20:10:11 -07:00
Ranran
dc6908ac6a [Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (#35007) 
Signed-off-by: Ranran <1012869439@qq.com>
Signed-off-by: Ranran <hzz5361@psu.edu>
Signed-off-by: ran <hzz5361@psu.edu>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
 2026-03-23 18:31:14 -04:00
yzong-rh
e85f8f0932 [Bug][MoE] Strengthen _supports_current_device() checks in the TRTLLM FP8, NVFP4, and FlashInfer CuteDSL MoE experts (#36728) 
Signed-off-by: Yifan Zong <yzong@redhat.com>
 2026-03-23 17:02:57 -04:00
Robert Shaw
5bf3c42d4c [Bug][MoE] Fix TRTLLM NVFP4 Routing Kernel Precision (#36725) 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
 2026-03-23 20:19:06 +00:00
Kyle Sayers
38364a7e32 [Sparse24] [Deprecation] Remove Sparse24 CT integration and kernels (#36799) 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
 2026-03-23 16:03:29 -04:00
Kunshang Ji
debd6e768c [XPU][MoE Refactor] Refactor xpu mxfp4 support into oracle (#37784) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-03-23 11:10:41 +00:00
Kunshang Ji
27d5ee3e6f [FP8]add FP8 WoQ kernel abstraction. (#32929) 
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
 2026-03-23 09:47:47 +00:00
Chuan (Richard) Li
e99fb98867 [ROCm] Fix fused_moe_fake signature mismatch and other AITER bugs (#36100) 
Signed-off-by: Li <chuali@amd.com>
 2026-03-23 15:48:31 +08:00
Artem Perevedentsev
a16133a0f1 [Perf] [Bugfix] Fix Triton autotuning in inference for Qwen3.5 (#37338) 
Signed-off-by: Artem Perevedentsev <aperevedents@nvidia.com>
 2026-03-23 00:37:58 -07:00
Matthias Gehre
410d300893 [ROCm][Refactor] Enable AWQMarlinConfig on ROCm to use choose_mp_linear_kernel (#36505) 
Signed-off-by: Matthias Gehre <matthias.gehre@amd.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-03-23 15:36:08 +08:00
Yongye Zhu
c058ff44d4 [Bigfix]fix lora test by pass padded size back to the layer (#37811) 2026-03-22 13:20:13 -06:00
Wentao Ye
77d24c4bfe [Bug] Fix fp8 deepgemm batch invariant (#37718) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-22 08:57:20 -04:00
Robert Shaw
4383f1532e [MoE] Move PF Methods to Folder (#35927) 2026-03-22 02:42:59 -06:00
Robert Shaw
6b2fa3a762 [MoE] Move FlashInfer CuteDSL experts into fused_moe/experts/ (#37759) 
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
 2026-03-21 19:15:16 -04:00
Robert Shaw
eeee5b262d [Quantization][Deprecation] Remove PTPC FP8 (#32700) 
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
 2026-03-21 22:10:16 +00:00
Robert Shaw
5ad0446572 Revert "Consolidate AWQ quantization into single awq_marlin.py file" (#37768) 2026-03-21 17:20:41 -04:00
Robert Shaw
8cc700dd6a Consolidate AWQ quantization into single awq_marlin.py file 
Merge awq.py and awq_marlin.py into a single file, eliminating the
circular import between them. awq.py becomes a backward-compat shim.
Follows the same structure as gptq_marlin.py.

Co-authored-by: Claude

Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
 2026-03-21 17:09:17 -04:00
Chaitanya Sri Krishna Lolla
3982bc2cd0 [ROCm] Enable DeepEP ROCm as all2allbackend for AMD GPUs. (#34692) 
Signed-off-by: Tej Kiran <vpolamre@amd.com>
Co-authored-by: Tej Kiran <vpolamre@amd.com>
 2026-03-21 00:32:31 -07:00
Yongye Zhu
87bd91892f [MoE Refactor] Mxfp4 oracle rebased (#37128) 
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 03:37:04 +00:00
Wentao Ye
b3d0b37908 [Refactor] Remove unused dead code (#36171) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-20 16:12:51 -06:00
Vadim Gimpelson
4f16ebbbd3 [Bugfix] Disable monolithic TRTLLM MoE for Renormalize routing (#37591) (#37605) 
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
 2026-03-20 12:19:26 -07:00
Xin Yang
d0532bf38d [Perf] Eliminate redundant SparseMatrix creation in gpt_oss_triton_kernels (#37683) 
Signed-off-by: Xin Yang <xyangx@amazon.com>
 2026-03-20 11:28:41 -06:00
L.B.R.
1779c09898 [ROCm] Enable wvSplitK skinny GEMM kernel for RDNA4/gfx1x decode (#34709) 
Signed-off-by: L.B.R. <lbr@mmonad.com>
Co-authored-by: L.B.R. <lbr@mmonad.com>
 2026-03-20 10:11:23 -05:00
xuebwang-amd
44eea10f68 [ROCm][Quantization] make quark ocp mx dtype parser robust for weight-only quantization (#36232) 
Signed-off-by: xuebwang-amd <xuebwang@amd.com>
 2026-03-20 10:10:03 -05:00
wang.yuqi
ed359c497a [Model] Deprecate the score task (this will not affect users). (#37537) 
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-03-20 08:07:56 +00:00
bnellnm
91be5f9be3 [MoE Refactor] Rename "naive" all2all backend (#36294) 
Signed-off-by: Bill Nell <bnell@redhat.com>
 2026-03-19 15:50:34 -04:00
bnellnm
9279c59a0e [MoE Refactor] DefaultMoERunner simplifcation (#33049) 
Signed-off-by: Bill Nell <bnell@redhat.com>
 2026-03-19 15:07:44 -04:00
Wei Zhao
e27b8ba3d1 [Bug] Fix fp8 trtllm MoE modular kernel supported routing methods (#37346) 
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
 2026-03-19 11:43:06 -04:00
Duyi-Wang
6a9cceb219 [Bugfix][ROCm] Fix MoRI + AITER FP8 dispatch compatibility for defer_input_quant (#37418) 
Signed-off-by: Duyi-Wang <duyi.wang@amd.com>
 2026-03-19 09:49:27 +00:00
Michael Goin
9482b0b085 [Bugfix] Remove assertion for NVFP4 scale dynamic range (#37465) 
Signed-off-by: Michael Goin <mgoin64@gmail.com>
 2026-03-18 15:37:49 -07:00
Wentao Ye
0d81a1fe61 [V0 Deprecation] Deprecate virtual engine (#37195) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-18 14:30:14 -07:00
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
Xin Yang
b1169d7be8 [Kernel] Add gpt-oss Router GEMM kernel (#37205) 
Signed-off-by: Xin Yang <xyangx@amazon.com>
 2026-03-18 08:15:56 -07:00
elvischenv
296839a1b0 [Perf] Eliminate padding and slicing op for GPT-OSS with Flashinfer MXFP4 MXFP8 MoE (#30647) 
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
 2026-03-18 15:01:26 +00:00
Wentao Ye
c373b5c00d [Log] Reduce duplicate log (#37313) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-03-18 10:57:44 -04:00
Michael Goin
09e4576f65 [Kernel] Add non-gated support for NVFP4 CUTLASS MoE (#37320) 
Signed-off-by: mgoin <mgoin64@gmail.com>
 2026-03-17 18:12:04 -04:00
Chao-Ju Chen
245758992e [Bugfix] Rescale NVFP4 weight scales to fix BF16 dequant underflow (#34577) 
Signed-off-by: ricky-chaoju <ricky.chen@infinirc.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-03-17 20:48:42 +00:00
Andrey Talman
68f783a727 [Torch 2.11] Guard torch._C._cpu attribute checks for forward compatibility (#35673) 
Signed-off-by: atalman <atalman@fb.com>
 2026-03-17 18:47:59 +00:00
Augusto Yao
9c7cab5ebb [Feature]: Support for multiple embedding types in a single inference call (#35829) 
Signed-off-by: augusto.yjh <augusto.yjh@antgroup.com>
 2026-03-17 17:05:42 +08:00