Commits · fccd5325874321f34daef0a8ed4d1b15b26ca34e · OpenDAS / vllm_cscc

09 Dec, 2025 8 commits

[Quantization] FP8 Weight Reloading for Quantized RL Rollout (#28480) · fccd5325
Kyle Sayers authored Dec 09, 2025
```
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
```
fccd5325

[MoE][Refactor] Remove most arguments to FusedMoEMethodBase.apply (#29066) · 00e5cbb9

bnellnm authored Dec 09, 2025

Signed-off-by: Bill Nell <bnell@redhat.com>
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>

00e5cbb9

[ROCm] Aiter Quant Kernels (#25552) · ee14644b
vllmellm authored Dec 09, 2025
```
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
```
ee14644b
[Bugfix]: Fix glm46 awq marlin moe wna16 compatibility (#30210) · 1166c31c
Dongjie Zou authored Dec 09, 2025
```
Signed-off-by: baonudesifeizhai <baonudesifeizhai@gmail.com>
```
1166c31c

[Model][Quantization] Restore MoE + GGUF models support (incl. Qwen3 MoE) by... · 58d5b3f5

Tsukasa OI authored Dec 09, 2025


[Model][Quantization] Restore MoE + GGUF models support (incl. Qwen3 MoE) by allowing Sideload Parameters (#30116)
Signed-off-by: Tsukasa OI <floss_llm@irq.a4lg.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

58d5b3f5

[Bugfix] Fix compressed-tensors models failing to load with transformers backend (#30287) · 03b91f72

Michael Goin authored Dec 08, 2025

Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

03b91f72

[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

[Bugfix] Fix DeepGEMM after #29546 (#30267) · ae339b1a

Zhewen Li authored Dec 08, 2025


Signed-off-by: zhewenli <zhewenli@meta.com>
Signed-off-by: Zhewen Li <zhewenli@meta.com>

ae339b1a

08 Dec, 2025 2 commits
- online fp8 quant with streaming weight post-processing (#29196) · 0d402d26
  Vasiliy Kuznetsov authored Dec 08, 2025
```
Signed-off-by: vasiliy <vasiliy@fb.com>
```
  0d402d26
- [ROCm] [Fused Moe EP] Use binary expert mask for aiter fused moe kernel (#29773) · c6df05eb
  Zhiwei authored Dec 08, 2025
```
Signed-off-by: ZhiweiYan-96 <zhiwei.yan@amd.com>
```
  c6df05eb
07 Dec, 2025 2 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

06 Dec, 2025 2 commits

Fix AWQ MoE marlin check issue in marlin_utils.py for AMD backend (#30102) · c4d62618
yuttian1 authored Dec 06, 2025
```
Signed-off-by: yuttian1 <yuttian@amd.com>
```
c4d62618

[CI/Build][AMD][Quantization] Fix test_int8_kernel.py by updating int8_utils... · dc839ad0

rasmith authored Dec 05, 2025


[CI/Build][AMD][Quantization] Fix test_int8_kernel.py by updating int8_utils to use hip.libdevice.round (#30151)
Signed-off-by: Randall Smith <ransmith@amd.com>
Co-authored-by: Randall Smith <ransmith@amd.com>

dc839ad0

05 Dec, 2025 2 commits
- [Bug] Fix vLLM config is not set error (#29999) · 7b5575fa
  Wentao Ye authored Dec 05, 2025
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
  7b5575fa
- [Compressed Tensors] Add XPU `wNa16` support (#29484) · 0d8a7d8a
  Yi Liu authored Dec 05, 2025
```
Signed-off-by: yiliu30 <yi4.liu@intel.com>
```
  0d8a7d8a
04 Dec, 2025 1 commit
- [Bugfix] Fix FP8 MoE LoRA (#29890) · 652ba93d
  Jee Jee Li authored Dec 05, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  652ba93d
03 Dec, 2025 3 commits

[Kernels] Remove BatchedTritonOrDeepGemmExperts and default fallback to Triton (#29929) · 2902c348

bnellnm authored Dec 03, 2025


Signed-off-by: Bill Nell <bnell@redhat.com>
Signed-off-by: bnellnm <49004751+bnellnm@users.noreply.github.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

2902c348

[Performance][DP/EP] Add silu_mul_per_token_group_quant_fp8_colmajor kernel (#29470) · 19bee6d1

Varun Sundar Rabindranath authored Dec 03, 2025

Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

19bee6d1

[refactor] CTMoEMethods to use QuantizationArgs (#28871) · b294e28d

HDCharles authored Dec 03, 2025


Signed-off-by: HDCharles <charlesdavidhernandez@gmail.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

b294e28d

01 Dec, 2025 1 commit
- [XPU] Fix AWQ skipped layer detection in IPEX quantization (#29774) · f37e8938
  Fanli Lin authored Dec 01, 2025
```
Signed-off-by: Fanli Lin <fanli.lin@intel.com>
```
  f37e8938
30 Nov, 2025 4 commits
- [Feat] Support non-gated activations in NVFP4 modelopt path (#29004) · 39d28108
  Omer Ullman Argov authored Nov 30, 2025
  
  39d28108
- [Bugfix] Fix mismatched nvfp4 gemm output shape (#29742) · 47539cfd
  Isotr0py authored Nov 30, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  47539cfd
- [Quantization] Enable compressed-tensors AWQ for Turing GPU (#29732) · e1464c3a
  Isotr0py authored Nov 30, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
  e1464c3a
- [LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 (#29708) · a491b091
  Xin Yang authored Nov 29, 2025
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
Signed-off-by: Xin Yang <105740670+xyang16@users.noreply.github.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  a491b091
29 Nov, 2025 2 commits

[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

[Doc]: fixing typos in various files. (#29717) · 04a797cd
Didier Durand authored Nov 29, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
04a797cd

28 Nov, 2025 2 commits
- Revert "[LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 (#28971)" (#29697) · 3fd1fb0b
  Huamin Li authored Nov 28, 2025
```
Signed-off-by: Huamin Li <3ericli@gmail.com>
```
  3fd1fb0b
- [LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 (#28971) · 745a3bae
  Xin Yang authored Nov 27, 2025
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  745a3bae
27 Nov, 2025 2 commits
- [Attention] Update attention imports (#29540) · fc1d8be3
  Matthew Bonanni authored Nov 27, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  fc1d8be3
- [Bugfix] Make compressed-tensors MoEs respect ignored layers (#28878) · df01eda4
  HDCharles authored Nov 26, 2025
```
Signed-off-by: HDCharles <charlesdavidhernandez@gmail.com>
```
  df01eda4
26 Nov, 2025 1 commit

[refactor] CTConfig methods to static/class methods (#28870) · e6031295

HDCharles authored Nov 26, 2025


Signed-off-by: HDCharles <charlesdavidhernandez@gmail.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

e6031295

25 Nov, 2025 3 commits
- Change warning logs to debug for unimplemented MXFP4 Linear/Attention (#29441) · 7df02897
  Michael Goin authored Nov 25, 2025
```
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
```
  7df02897
- [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
- [Perf] Disable DeepGEMM MoE by default when TP=8 is used (#29346) · 6f1355a1
  Michael Goin authored Nov 24, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  6f1355a1
24 Nov, 2025 2 commits
- [Bugfix] Only use triton_kernels for MXFP4 on SM90 and SM100 (#29339) · c17610e2
  Michael Goin authored Nov 24, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  c17610e2
- [MoE][Refactor] Make select_experts a non-static method (#29067) · 8f066146
  bnellnm authored Nov 24, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  8f066146
23 Nov, 2025 1 commit

[Fix] Add SM check to flashinfer MOE backend (#29144) · 5f96c00c

jiahanc authored Nov 22, 2025


Signed-off-by: jiahanc <173873397+jiahanc@users.noreply.github.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>

5f96c00c

22 Nov, 2025 1 commit
- [Model][Qwen3VL] Tune Triton w8a8 block fp8 kernel for L40s (#29217) · d045e22d
  Lukas Geiger authored Nov 22, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  d045e22d
21 Nov, 2025 1 commit

Add TRTLLM MoE NVFP4 kernel to CompressedTensorsW4A4MoeMethod (#28892) · b4c8fbaa

Mingyuan Ma authored Nov 21, 2025


Signed-off-by: mingyuanm <mingyuanm@nvidia.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>

b4c8fbaa