Commits · 2f4bdee61ee0dd9358efaba720b7acc53b2ece00 · OpenDAS / vllm_cscc

06 Jan, 2026 1 commit
- [Quantization][MoE] remove unused ep logic from moe marlin (#31571) · 2f4bdee6
  Jinzhen Lin authored Jan 07, 2026
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
  2f4bdee6
29 Dec, 2025 1 commit
- [ROCm][GPTQ][Bugfix] Fix GPTQ GEMM kernel output zeroing race condition (#30719) · 3ecfdc37
  Andreas Karatzas authored Dec 29, 2025
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  3ecfdc37
24 Dec, 2025 3 commits
- [PERF] Add interleaved memory allocation to NUMA module (#30800) · 7cd288a4
  skaraban3807 authored Dec 24, 2025
  
  7cd288a4
- [DeepSeek v3.2] Remove unnecessary syncwarps (#31047) · 369f47aa
  Matthew Bonanni authored Dec 24, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  369f47aa
- [Misc] Remove unused custom ops `copy_blocks` and `copy_blocks_mla` (#30967) · 4ed11105
  rongfu.leng authored Dec 24, 2025
```
Signed-off-by: rongfu.leng <rongfu.leng@daocloud.io>
```
  4ed11105
23 Dec, 2025 1 commit

[FIX] FP4 quantization kernel padding initialization bug (#31097) · b94f80ff

danielafrimi authored Dec 23, 2025



Signed-off-by: <>
Co-authored-by: root <root@gpu-193.slurm-workers-slurm.slurm.svc.cluster.local>
Co-authored-by: root <root@gpu-951.slurm-workers-slurm.slurm.svc.cluster.local>

b94f80ff

22 Dec, 2025 2 commits
- [ROCm] [Critical]: Remove unused variable (#31156) · 022f3cea
  TJian authored Dec 23, 2025
```
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
```
  022f3cea
- [Kernel] Enable fused_qknorm_rope_kernel supports partial rope (#30821) · 097978a1
  Jee Jee Li authored Dec 22, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  097978a1
21 Dec, 2025 1 commit
- [NVFP4][Perf] Tune NVFP4 input quant kernel for small batch size (#30897) · 06d49028
  Michael Goin authored Dec 21, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  06d49028
19 Dec, 2025 2 commits
- [MoE Refactor][3/N] Deprecate cutlass block quant fp8 (b200) (#30990) · 83a317f6
  Robert Shaw authored Dec 19, 2025
```
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
```
  83a317f6
- [CPU][Bugfix] Fix ppc64le CPU build (#30871) · bd2b52fc
  Nishidha Panpaliya authored Dec 19, 2025
```
Signed-off-by: Nishidha Panpaliya <nishidha.panpaliya@partner.ibm.com>
```
  bd2b52fc
18 Dec, 2025 2 commits
- [Bugfix][CPU] Fix Mac CPU build (#30955) · f90d3636
  Li, Jiang authored Dec 18, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  f90d3636
- [CPU] Refactor CPU fused MOE (#30531) · e3ab93c8
  Li, Jiang authored Dec 18, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  e3ab93c8
17 Dec, 2025 1 commit
- [NIXL][Bugfix] Fix NIXL/RDMA registration failure over CuMemAllocator (#29569) · f4e884f2
  Sheng Lin authored Dec 17, 2025
```
Signed-off-by: Somoku <linsh0@protonmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
```
  f4e884f2
16 Dec, 2025 4 commits

[Perf][Kernels] Vectorize `csrc/activations_kernels.cu` (#29512) · 0a1ab1e5
Michael Goin authored Dec 16, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
0a1ab1e5
[Kernel][Quantization][MoE] add marlin kernel support for turing (sm75) (#29901) · ce96857f
Jinzhen Lin authored Dec 17, 2025
```
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
ce96857f

[Bugfix][DSV32] Fix overflow in topk. (#30754) · eaa82a70

Daniel Cámpora authored Dec 16, 2025


Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>

eaa82a70

[Refactor] Small refactor for group topk (#30562) · f21f5ea3

Wentao Ye authored Dec 16, 2025


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

f21f5ea3

12 Dec, 2025 3 commits
- [Refactor] Reduce duplicate code in `per_token_group_quant` cuda kernels (#30496) · 1e6b1153
  Wentao Ye authored Dec 12, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  1e6b1153
- [Attention] Use sparse prefill kernel for fp8 kv-cache in DeepSeek-v3.2 (#27532) · 3e41992f
  Lucas Wilkinson authored Dec 12, 2025
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  3e41992f
- gptq marlin quantization support for fused moe with lora (#30254) · 6a6fc41c
  Bhanu Prakash Voutharoja authored Dec 12, 2025
```
Signed-off-by: Bhanu068 <voutharoja.bhanu06@gmail.com>
```
  6a6fc41c
11 Dec, 2025 1 commit
- [Refactor] Remove useless syncwarp (#30510) · 61249b17
  Wentao Ye authored Dec 11, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  61249b17
10 Dec, 2025 2 commits
- [CPU] Support for Whisper (#30062) · cebda2a4
  Aditya Tewari authored Dec 10, 2025
```
Signed-off-by: Aditya Tewari <aditya.tewari@arm.com>
```
  cebda2a4
- Fix typos in comments across multiple files (#30345) · 3bdd4266
  Wilson Wu authored Dec 10, 2025
```
Signed-off-by: Wilson Wu <iwilsonwu@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
```
  3bdd4266
09 Dec, 2025 4 commits
- Improve wvsplitK tile and balance heristics. (#29937) · 2e7054da
  Hashem Hashemi authored Dec 09, 2025
```
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
```
  2e7054da
- [Kernel]Support W4A8 Grouped GEMM on Hopper (#29691) · f6227c22
  czhu-cohere authored Dec 08, 2025
```
Signed-off-by: czhu-cohere <conway.zhu@cohere.com>
```
  f6227c22
- Lora MoE Align Improvements (#29257) · ea657f20
  gnovack authored Dec 08, 2025
```
Signed-off-by: gnovack <gnovack@amazon.com>
```
  ea657f20
- [Perf] Optimize `group_topk` kernel, 1.9% Throughput improvement, 2.1% TPOT improvemnt (#30159) · 0ee6416f
  Wentao Ye authored Dec 08, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  0ee6416f
08 Dec, 2025 1 commit

[DeepSeek v3.2] Make top-k work for any logit values. (#27568) · 184076c3

Daniel Cámpora authored Dec 08, 2025


Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

184076c3

07 Dec, 2025 3 commits

[Performance] Fused blockwise quant RMS norm (#27883) · af0444bf

ElizaWszola authored Dec 07, 2025


Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: yewentao256 <zhyanwentao@126.com>

af0444bf

[Perf] Deepgemm fused layout kernel for activations, 4.3% throughput... · 541a2ef8

Wentao Ye authored Dec 07, 2025


[Perf] Deepgemm fused layout kernel for activations, 4.3% throughput improvement, 10.7% TTFT improvement. (#29546)
Signed-off-by: yewentao256 <zhyanwentao@126.com>

541a2ef8

[Kernel][MoE] optimize `moe_align_block_size` (#29642) · 879ddb09

Jinzhen Lin authored Dec 07, 2025


Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>

879ddb09

05 Dec, 2025 1 commit

[CPU][Perf] Add fast vectorized exp impl from Arm Optimized Routines (#30068) · 9843e332

Elham authored Dec 05, 2025

Signed-off-by: Ubuntu <ubuntu@ip-10-252-30-150.eu-west-1.compute.internal>
Signed-off-by: Elham Harirpoush <elham.harirpoush@arm.com>
Co-authored-by: Ubuntu <ubuntu@ip-10-252-30-150.eu-west-1.compute.internal>

9843e332

02 Dec, 2025 2 commits
- [CPU]Parallelize over tokens in int4 moe (#29600) · 13ea39bc
  Zhang Xiangze authored Dec 02, 2025
```
Signed-off-by: Zhang Xiangze <Xiangze.Zhang@arm.com>
```
  13ea39bc
- SM120 / NVFP4: add device guard and runtime SM dispatch to cutlass_scaled_fp4_mm (#29711) · c0dfc894
  Hendrik Holtmann authored Dec 02, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
```
  c0dfc894
29 Nov, 2025 1 commit

[Kernel][Quantization] add w4a8 support for marlin kernel (#24722) · 1656ad37

Jinzhen Lin authored Nov 29, 2025


Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin@redhat.com>

1656ad37

28 Nov, 2025 1 commit
- [CPU] Update torch 2.9.1 for CPU backend (#29664) · e2f56c30
  Li, Jiang authored Nov 28, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  e2f56c30
27 Nov, 2025 1 commit

[Bugfix] fix IMA issue in certain cases of the moe marlin kernel (#28619) · a67dec7c

Jinzhen Lin authored Nov 27, 2025


Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

a67dec7c

26 Nov, 2025 1 commit
- [Performance][MLA][ROCm] Remove redundant D2D copy in deepseek (#27457) · d9d342d2
  Pleaplusone authored Nov 26, 2025
```
Signed-off-by: ganyi <ygan@amd.com>
```
  d9d342d2
25 Nov, 2025 1 commit
- [Kernel] Add NVFP4 MoE CUTLASS support for SM120 (#29242) · e5020986
  Michael Goin authored Nov 25, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
```
  e5020986