Commits · dbebb7f812123b4a0efe5b085582d3345fe7f740 · OpenDAS / vllm_cscc

16 Sep, 2025 9 commits

[Perf] Reuse workspace for FP8+FP4 Marlin MoE (#20500) · dbebb7f8

Michael Goin authored Sep 16, 2025


Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

dbebb7f8

fp8 kv cache support fix for torch.compile (#22758) · 3053a22b

Aleksandr Malyshev authored Sep 16, 2025

Signed-off-by: Aleksandr Malyshev <maleksan@amd.com>
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Co-authored-by: Aleksandr Malyshev <maleksan@amd.com>
Co-authored-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Co-authored-by: Gregory Shtrasberg <156009573+gshtras@users.noreply.github.com>

3053a22b

[Core/DBO][1/N] Add Dual-Batch Overlap mechanism to VLLM (#23693) · 56793995

Sage Moore authored Sep 16, 2025


Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Sage Moore <sage@neuralmagic.com>
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
Co-authored-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

56793995

[Model] Clean up and simplify Mamba2 Metadata Usage in both V0 and V1 (#24331) · 73cfb3c5
Chih-Chieh Yang authored Sep 16, 2025
```
Signed-off-by: Chih-Chieh-Yang <7364402+cyang49@users.noreply.github.com>
```
73cfb3c5

[Feat][EPLB] A novel static EPLB placement strategy for MoE models. (#23745) · 7ea5c73a

Chen Bruce authored Sep 16, 2025


Signed-off-by: bruceszchen <bruceszchen@tencent.com>
Signed-off-by: Chen Bruce <bruceszchen@tencent.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Chen Bruce <cszwwdz@vip.qq.com>
Co-authored-by: lemon412 <lemon412@foxmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

7ea5c73a

[Mamba] Support TP>1 with quantization for mamba2 mixer in case `n_groups % tp_size == 0` (#24593) · 27fcfe7b

tomeras91 authored Sep 16, 2025

Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
Signed-off-by: tomeras91 <57313761+tomeras91@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

27fcfe7b

[benchmark] Add triton version in the moe tuned config (#24769) · 04ad0dc2
Jee Jee Li authored Sep 16, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
04ad0dc2
[QWEN NEXT] Fused MoE kernels Optimization configs (#24924) · 238c4c17
Saman A. Pour authored Sep 15, 2025
```
Signed-off-by: Saman Keon <samanamp@outlook.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
238c4c17
[ROCm][Bugfix] Fix the case where there's bias (#24895) · 2891603e
Gregory Shtrasberg authored Sep 15, 2025
```
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
```
2891603e

15 Sep, 2025 3 commits
- [Transform] Deterministic Hadacore Transforms (#24106) · a0b26701
  Kyle Sayers authored Sep 15, 2025
```
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
```
  a0b26701
- [USAGE] Improve error handling for weight initialization in Unquantized… (#20321) · b834b4cb
  Rafael Marcelino Koike authored Sep 15, 2025
```
Signed-off-by: Rafael Marcelino Koike <rafael.koike@oracle.com>
Signed-off-by: Rafael Koike <koike.rafael@gmail.com>
```
  b834b4cb
- [Doc]: fix typos in various files (#24821) · 4979eb79
  Didier Durand authored Sep 15, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
  4979eb79
14 Sep, 2025 1 commit

[Perf] Fix DeepGEMM Contiguous Layout Issue, 5.5% Throughput Improvement (#24783) · fc2dbcda

Wentao Ye authored Sep 14, 2025


Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

fc2dbcda

13 Sep, 2025 3 commits
- [Doc]: fix typos in various files (#24798) · 41ae4a1e
  Didier Durand authored Sep 13, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
  41ae4a1e
- [Kernels][DP/EP] Optimize Silu Kernel for R1 (#24054) · 98229db2
  Elvir Crnčević authored Sep 13, 2025
```
Signed-off-by: elvircrn <elvircrn@gmail.com>
```
  98229db2
- [Chore] Remove unused batched RoPE op & kernel (#24789) · 5febdc87
  Woosuk Kwon authored Sep 13, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  5febdc87
12 Sep, 2025 8 commits
- [Attention][FlashInfer] Enable FP8 FlashInfer (TRTLLM) MLA decode (#24705) · 7ba32aa6
  Matthew Bonanni authored Sep 12, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
```
  7ba32aa6
- [Qwen3-Next] MoE configs for H100 TP=1,2 and TP2/EP (#24739) · 9f04d9d5
  Elvir Crnčević authored Sep 12, 2025
```
Signed-off-by: elvircrn <elvircrn@gmail.com>
```
  9f04d9d5
- [Bugfix] Fix MRoPE dispatch on XPU (#24724) · 4d7c1d53
  Yan Ma authored Sep 12, 2025
```
Signed-off-by: Yan Ma <yan.ma@intel.com>
```
  4d7c1d53
- [Docs] Fix warnings in mkdocs build (continued) (#24740) · 41f17bf2
  Hyogeun Oh (오효근) authored Sep 12, 2025
```
Signed-off-by: Zerohertz <ohg3417@gmail.com>
```
  41f17bf2
- [Doc]: fix typos in various files (#24726) · bcb06d7b
  Didier Durand authored Sep 12, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
  bcb06d7b
- [Bugfix] Fix MRoPE dispatch on CPU (#24712) · 7920de0a
  Li, Jiang authored Sep 12, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  7920de0a
- [Qwen3-Next] MoE configs for H20 TP=1,2,4,8 (#24707) · 12a8414d
  Jee Jee Li authored Sep 12, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  12a8414d
- [Bugfix] fixes the causal_conv1d_update kernel update non-speculative decoding cases (#24680) · 880c741b
  Tao He authored Sep 12, 2025
```
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  880c741b
11 Sep, 2025 15 commits
- [Bug] Fix Layer `weight_block_size` Assertion Issue (#24674) · fcba05c4
  Wentao Ye authored Sep 11, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  fcba05c4
- [Qwen3-Next] MOE configs for H100 TP4 (#24699) · f82f7a89
  Chen Zhang authored Sep 11, 2025
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  f82f7a89
- [Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280) · c3aea10d
  Michael Goin authored Sep 11, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
```
  c3aea10d
- [Qwen3-Next] Add B200 MoE configs for Qwen3-next (#24698) · 7a70a718
  Vadim Gimpelson authored Sep 12, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
```
  7a70a718
- [Qwen3-Next] MoE configs for H200 TP=1,2,4 (#24695) · 569bf1c9
  Woosuk Kwon authored Sep 11, 2025
```
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
```
  569bf1c9
- [Kernel][B200] `mxfp4` fused cutlass moe (#23696) · 074854b2
  Duncan Moss authored Sep 11, 2025
```
Signed-off-by: Duncan Moss <djm.moss@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  074854b2
- [Qwen3-Next] Add MoE Config for H200 (#24688) · c733bd5e
  Woosuk Kwon authored Sep 11, 2025
```
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
```
  c733bd5e
- [Doc] Remove Useless Comments (#24687) · a892b259
  Wentao Ye authored Sep 11, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  a892b259
- fix some typos (#24616) · e26fef83
  co63oc authored Sep 12, 2025
```
Signed-off-by: co63oc <co63oc@users.noreply.github.com>
```
  e26fef83
- [Bugfix] Fix platform-specific routing in CustomOp implementations (#24444) · 4aa23892
  Konrad Zawora authored Sep 11, 2025
```
Signed-off-by: Konrad Zawora <kzawora@habana.ai>
```
  4aa23892
- Add the support for the qwen3 next model (a hybrid attention model). (#24526) · e93f4cc9
  Tao He authored Sep 11, 2025
```
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  e93f4cc9
- [torchao] Support quantization configs using module swap (#21982) · 2048c4e3
  Jerry Zhang authored Sep 10, 2025
```
Signed-off-by: Jerry Zhang <jerryzh168@gmail.com>
```
  2048c4e3
- [Model] New model support for Motif-1-Tiny (#23414) · 9bd831f5
  TaehyunKim authored Sep 11, 2025
```
Signed-off-by: ca1207 <ca1207zzz@gmail.com>
Signed-off-by: TaehyunKim <73943231+ca1207@users.noreply.github.com>
Co-authored-by: WyldeCat <skan1543@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  9bd831f5
- [Doc]: fixing doc typos (#24635) · e2b1f863
  Didier Durand authored Sep 11, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
  e2b1f863
- Kimi K2 Fused MoE kernels Optimization configs (#24597) · 3d1393f6
  Saman A. Pour authored Sep 10, 2025
```
Signed-off-by: Saman Keon <samanamp@outlook.com>
```
  3d1393f6
10 Sep, 2025 1 commit

[torch.compile][ROCm][V1] Enable attention output FP8 fusion for V1 attention backends (#19767) · 9a161307

Gregory Shtrasberg authored Sep 10, 2025


Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

9a161307