Commits · 63400259d05485330fb23635af892b3fce160dea · OpenDAS / vllm_cscc

23 Sep, 2025 9 commits

[Performance] Move apply_w8a8_block_fp8_linear to an op class (#24666) · 63400259

ElizaWszola authored Sep 23, 2025


Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: ElizaWszola <elizaw.9289@gmail.com>
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedič <lgovedic@redhat.com>

63400259

[ROCm] Add skinny gemm bias support for dtypes fp16,bf16,fp8 (#24988) · a3a78280

Hashem Hashemi authored Sep 23, 2025


Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
Signed-off-by: Hashem Hashemi <159079214+amd-hhashemi@users.noreply.github.com>

a3a78280

[Kernels] Support blocked fp8 quantization for compressed tensors MoE (#25219) · f11e3c51
bnellnm authored Sep 23, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
```
f11e3c51

[Core/DBO][2/N] Dual-Batch Overlap add DeepEP High Throughput support and Prefill support (#24845) · cc1dc7ed

Lucas Wilkinson authored Sep 23, 2025


Signed-off-by: Sage Moore <sage@neuralmagic.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Sage Moore <sage@neuralmagic.com>
Co-authored-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

cc1dc7ed

[V1] Remove V0 code paths for Hybrid models (#25400) · a903669e
Thomas Parnell authored Sep 23, 2025
```
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
```
a903669e

[BugFix] Register expert_map as named buffer for wake_up and sleep (#25458) · 61d1b355

Joel authored Sep 23, 2025


Signed-off-by: wuxibin <wuxibin@bytedance.com>
Signed-off-by: youkaichao <youkaichao@gmail.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>

61d1b355

Handle triton kernel import exception (#25319) · ba8d2165
Ming Yang authored Sep 23, 2025
```
Signed-off-by: Ming Yang <minos.future@gmail.com>
```
ba8d2165
[DP/EP][GPTOSS] Use triton matmul-ogs kernels for GPTOSS DP/EP (#24588) · e8db44f8
Varun Sundar Rabindranath authored Sep 23, 2025
```
Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
```
e8db44f8

[feat] Support MRoPE + YaRN (#25384) · fc97733d

JJJYmmm authored Sep 23, 2025


Signed-off-by: liuye.hj <liuye.hj@alibaba-inc.com>
Co-authored-by: liuye.hj <liuye.hj@alibaba-inc.com>

fc97733d

22 Sep, 2025 2 commits

[EPLB] Reduce EPLB Inference Overhead (#24573) · 06a41334

Bowen Wang authored Sep 22, 2025

Signed-off-by: Bowen Wang <abmfy@icloud.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

06a41334

[Kernel] MI-300X triton moe configs (#23445) · ac243886
Sara-KS authored Sep 22, 2025
```
Signed-off-by: Sara Kokkila Schumacher <saraks@ibm.com>
```
ac243886

21 Sep, 2025 3 commits
- Remove V0 attention backends (#25351) · bc6e542d
  Woosuk Kwon authored Sep 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
  bc6e542d
- [V0 Deprecation] Remove V0 sampling metadata (#25345) · 1c3ffdbe
  Woosuk Kwon authored Sep 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
```
  1c3ffdbe
- [V0 Deprecation] Remove V0 Sequence class & Sampler (#25332) · 26e673fe
  Woosuk Kwon authored Sep 21, 2025
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>
```
  26e673fe
20 Sep, 2025 1 commit
- [CI Failure] Disable FlashInfer RoPE to unblock CI (#25299) · e08a3a3f
  Michael Goin authored Sep 20, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  e08a3a3f
19 Sep, 2025 9 commits

Enable modelopt gemma3 nvfp4/fp8, make workflow more robust (#22771) · 431535b5

Zhiyu authored Sep 19, 2025


Signed-off-by: Zhiyu Cheng <zhiyuc@nvidia.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>

431535b5

Fix: Correct FusedMoE layer reference in auto_round quantization (#24818) · ddc90483

David-Wen authored Sep 20, 2025

Signed-off-by: David-Wen <18927700430@163.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

ddc90483

[Perf] Use FlashInfer RoPE for RotaryEmbedding.forward_cuda when available (#21126) · 48ecb443

Michael Goin authored Sep 19, 2025


Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

48ecb443

[Bugfix] Fix chunked a2_scales in modular kernels (#25264) · 4bdf4002
bnellnm authored Sep 19, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
4bdf4002

[Bugfix] GPT OSS Attritbute error on H100 (#25228) · 7852b82b

Varun Sundar Rabindranath authored Sep 19, 2025


Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>

7852b82b

[Docs] add __init__.py to... · 138f0d1e

samzong authored Sep 20, 2025


[Docs] add __init__.py to vllm/model_executor/layers/quantization/compressed_tensors/transform (#24974)
Signed-off-by: samzong <samzong.lu@gmail.com>

138f0d1e

[Kernel][Performance] Add Triton kernel for Qwen3-VL interleaved MRoPE (#25055) · cea91a32
Isotr0py authored Sep 19, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
cea91a32

[Bugfix][CPU] Add placeholder to avoid import errors when using fused_moe ops... · 825fdb11

Li, Jiang authored Sep 19, 2025


[Bugfix][CPU] Add placeholder to avoid import errors when using fused_moe ops on platforms without triton (#25137)
Signed-off-by: jiang1.li <jiang1.li@intel.com>

825fdb11

[CPU] Disable oneDNN linear on non-x86 platforms (#25166) · 8c1d4acb
Li, Jiang authored Sep 19, 2025
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
8c1d4acb

18 Sep, 2025 7 commits
- [PERF] Add `conv1d` metadata to GDN attn (#25105) · 072d7e53
  Vadim Gimpelson authored Sep 18, 2025
```
Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
```
  072d7e53
- Refactor dense FP8 tensor/channel/block utils and add CT FP8 block (#21404) · fbd6523a
  Michael Goin authored Sep 18, 2025
  
  fbd6523a
- [Bugfix][Mamba] - Fix Conv State Kernel FP32 Support (#24883) · 66072b36
  Asaf Joseph Gardin authored Sep 18, 2025
```
Signed-off-by: asafg <39553475+Josephasafg@users.noreply.github.com>
```
  66072b36
- [Model] Improve Pooling Model (#25149) · 37970105
  Jee Jee Li authored Sep 18, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  37970105
- [Doc] Fix cross-reference warnings (#25058) · 05b044e6
  Punitvara authored Sep 18, 2025
```
Signed-off-by: Punit Vara <punitvara@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  05b044e6
- [EPLB] Add EPLB support for hunyuan_v1 (#23078) · 9d8a2d86
  YiwenC authored Sep 17, 2025
  
  9d8a2d86
- [Kernels] Overlap shared experts with combine instead of dispatch (#24254) · dc2979c5
  bnellnm authored Sep 18, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
  dc2979c5
17 Sep, 2025 5 commits

[Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537) · 5963b98b
bnellnm authored Sep 17, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
5963b98b
[Bugfix][Qwen3-Next] fixes the varlen issue in qwen3-next's MTP implementation. (#24957) · dd6a910a
Tao He authored Sep 17, 2025
```
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
```
dd6a910a
[Rocm] [quantization] Fix quark ptpc moe and add test case (#24649) · ca2d1925
haoyangli-amd authored Sep 17, 2025
```
Signed-off-by: Haoyang Li <lihaoyang0109@gmail.com>
Co-authored-by: Haoyang Li <haoyang.li@amd.com>
```
ca2d1925

[Model] Support Qwen3-VL Model Series (#24727) · 0f7acdd7

Roger Wang authored Sep 16, 2025


Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

0f7acdd7

[FP8] Extend per-token-group quantization support to QuantFP8 (#24342) · cef32104

Tahsin Tunan authored Sep 17, 2025

Signed-off-by: Tahsin Tunan <tahsintunan@gmail.com>
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Luka Govedič <lgovedic@redhat.com>

cef32104

16 Sep, 2025 4 commits

[CI][Bugfix] Fix failing Blackwell test (#24993) · d119fc86

Matthew Bonanni authored Sep 16, 2025


Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

d119fc86

[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