Commits · 302ef403a2305e9158064f8e386d1b5284d12cb2 · OpenDAS / vllm_cscc

15 Oct, 2025 4 commits

[DSA][MLA] Tiny refactor on DeepSeek to make it reusable for different backends (#26656) · 302ef403
Mengqing Cao authored Oct 15, 2025
```
Signed-off-by: MengqingCao <cmq0113@163.com>
```
302ef403
[Bugfix][Multi Modal] Fix incorrect Molmo token processing (#26873) · 8865da15
sangho.lee authored Oct 15, 2025
```
Signed-off-by: sanghol <sanghol@allenai.org>
```
8865da15
[Bugfix] Fix qwen3-omni audio truncation issue (#26815) · 8c851f6d
Isotr0py authored Oct 15, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
8c851f6d

[Frontend][torch.compile] CompilationConfig Overhaul (#20283): name change ... · 96b9aa5a

Morrison Turnansky authored Oct 14, 2025


[Frontend][torch.compile] CompilationConfig Overhaul (#20283): name change  compilation level to compilation mode, deprecation compilation level (#26355)
Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

96b9aa5a

14 Oct, 2025 12 commits
- [CI Failure] Fix torchao dep failure for Quantization Test (#26824) · 7e0ef408
  Michael Goin authored Oct 14, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  7e0ef408
- Added MoE configs for llama 4, H200 device with tp=4/8 tuning (#26837) · 0e658189
  Dhruvil Bhatt authored Oct 14, 2025
```
Signed-off-by: Dhruvil Bhatt <bhattdbh@amazon.com>
```
  0e658189
- llama4_vision_rope: add HIP override to accept (q, k) and avoid (positions, q, k) mismatch (#26790) · 87efc681
  Huamin Li authored Oct 14, 2025
```
Signed-off-by: Huamin Li <3ericli@gmail.com>
```
  87efc681
- [Kernel][MoE] Add MoE tunings for GLM 4.6-FP8 and GLM 4.5 Air on NVidia B200 (#26818) · aba48f7d
  Ze'ev Klapow authored Oct 14, 2025
  
  aba48f7d
- Adjusted the model order of the model registration file (#26798) · e9f1b8c9
  汪志鹏 authored Oct 14, 2025
```
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
```
  e9f1b8c9
- [Chore] Remove `SupportsV0Only` interface and update supported models docs (#26783) · 9c4cb683
  Cyrus Leung authored Oct 14, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  9c4cb683
- [Model] Use merge_by_field_config for MM models (O-P) (#26776) · 74704d45
  Cyrus Leung authored Oct 14, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  74704d45
- [Bugfix] Standardize merging multimodal embeddings (#26771) · d2f816d6
  Cyrus Leung authored Oct 14, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d2f816d6
- [Model][Bugfix]fix ernie45 load failed due to ernie45 eplb code (#26684) · 01ad27fa
  CSWYF3634076 authored Oct 14, 2025
```
Signed-off-by: wangyafeng <wangyafeng@baidu.com>
```
  01ad27fa
- [Feature][Quantization] auto_round format add support for regex (#24024) · 29350922
  Heng Guo authored Oct 14, 2025
```
Signed-off-by: n1ck-guo <heng.guo@intel.com>
Signed-off-by: Heng Guo <heng.guo@intel.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
```
  29350922
- [torch.compile] Unwrap fused_marlin_moe custom op (#26739) · 8ae16928
  Varun Sundar Rabindranath authored Oct 13, 2025
```
Signed-off-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
Co-authored-by: Varun Sundar Rabindranath <vsundarr@redhat.com>
```
  8ae16928
- [UX] Replace VLLM_ALL2ALL_BACKEND with --all2all-backend (#26732) · 3e051bda
  Michael Goin authored Oct 13, 2025
```
Signed-off-by: mgoin <mgoin64@gmail.com>
```
  3e051bda
13 Oct, 2025 13 commits

[FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate... · e3fdb627

Morrison Turnansky authored Oct 13, 2025


[FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate use_inductor in favor of backend, simplify custom_ops  (#26502)
Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@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: Jiangyun Zhu <riverclouds.zhu@qq.com>

e3fdb627

[Bug] Fix Assertion error DeepEP/csrc/kernels/intranode.cu:928: 'false and... · 7200a21c

Wentao Ye authored Oct 13, 2025


[Bug] Fix Assertion error DeepEP/csrc/kernels/intranode.cu:928: 'false and Unsupported type' (#26532)
Signed-off-by: yewentao256 <zhyanwentao@126.com>

7200a21c

[Quantization] [Performance] Enable Marlin GEMM kernels for the... · 89342ce4

Alex Kogan authored Oct 13, 2025


[Quantization] [Performance] Enable Marlin GEMM kernels for the calibration-free RTN-based quantization (#26051)
Signed-off-by: Alex Kogan <alex.kogan@oracle.com>
Signed-off-by: Alex Kogan <82225080+sakogan@users.noreply.github.com>

89342ce4

[Model] Use merge_by_field_config for MM models (M-N) (#26710) · afc47e4d
Cyrus Leung authored Oct 14, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
afc47e4d
[Bugfix][Speculative Decoding] Extend Eagle quantization config fix to llama_eagle.py (#26590) · e3b90c1b
Rahul Tuli authored Oct 13, 2025
```
Signed-off-by: Rahul Tuli <rtuli@redhat.com>
```
e3b90c1b
[CI/Build] upgrade compressed-tensors to 0.12.2 to address LGPLv3 (#26501) · a1b2d658
Sangyeon Cho authored Oct 14, 2025
```
Signed-off-by: Sangyeon Cho <josang1204@gmail.com>
```
a1b2d658

[UX] Speedup DeepGEMM warmup with heuristics (#25619) · 0d21b9b5

Michael Goin authored Oct 13, 2025

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

0d21b9b5

[unrevert] Add batch invariant kernel override for FlashInfer backend [2/n] (#26373) · 32637990
Bram Wasti authored Oct 13, 2025
```
Signed-off-by: Bram Wasti <bwasti@meta.com>
Signed-off-by: Bram Wasti <bwasti@fb.com>
```
32637990
[Bugfix] Fix out of bound index issue for Jina-embedding-v3 RoPE with cuda graph (#26687) · 8e67b255
Isotr0py authored Oct 13, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
8e67b255
[Model][0/N] Improve all pooling task | clean up (#25817) · 767c3ab8
wang.yuqi authored Oct 13, 2025
```
Signed-off-by: wang.yuqi <noooop@126.com>
```
767c3ab8
[Model] Fix Skywork R1V mlp (#26673) · 98f30b8c
Jee Jee Li authored Oct 13, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
98f30b8c
[FIX] Throwing an exception when the model does not support pool tasks (#25840) (#25855) · 46ad7395
yyzxw authored Oct 13, 2025
```
Signed-off-by: zxw <1020938856@qq.com>
Co-authored-by: wang.yuqi <noooop@126.com>
```
46ad7395
[Misc] cache result of disable_inplace (#26666) · 60e419c1
bnellnm authored Oct 12, 2025
```
Signed-off-by: Bill Nell <bnell@redhat.com>
```
60e419c1

12 Oct, 2025 7 commits

[easy] fix pre commit error on trunk (#26665) · 4fca1a1b
Huamin Li authored Oct 12, 2025
```
Signed-off-by: Huamin Li <3ericli@gmail.com>
```
4fca1a1b
[Models][Qwen3VL] Speedup `fast_pos_embed_interpolate` (#26647) · a6049be7
Lukas Geiger authored Oct 12, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
a6049be7
[Feature] Add support for naver/splade-v3 (BERT-based sparse embedding model) (#26339) · 18ed7746
gjgjos authored Oct 13, 2025
```
Signed-off-by: gjgjos <gjgjos@naver.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
18ed7746
Update `Optional[x]` -> `x | None` and `Union[x, y]` to `x | y` (#26633) · 8fcaaf6a
Harry Mellor authored Oct 12, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
8fcaaf6a
[Bugfix][CI/Build] Fix failing Mteb CI (#26638) · 045b396d
Isotr0py authored Oct 12, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
045b396d

[PERF] [Qwen3-next] Speed up gated RMSNorm (#26207) · 82e64c7a

Vadim Gimpelson authored Oct 12, 2025

Signed-off-by: Vadim Gimpelson <vadim.gimpelson@gmail.com>
Signed-off-by: Vadim Gimpelson <156319763+vadiklyutiy@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

82e64c7a

[EPLB] Support ernie4.5-moe (#22100) · c5c8f5ea

Haisheng Chen authored Oct 11, 2025


Signed-off-by: Haisheng Chen <langzs335@outlook.com>
Signed-off-by: Haisheng Chen <60504847+HsChen-sys@users.noreply.github.com>
Signed-off-by: Haisheng Chen <hac048@ucsd.edu>
Co-authored-by: Haisheng Chen <langzs335@outlook.com>

c5c8f5ea

11 Oct, 2025 4 commits

[Bugfix] Fix qwen-moe packed_modules_mapping (#26634) · f0a30a06
Jee Jee Li authored Oct 11, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
f0a30a06

[Bugfix][Qwen3VL] fix deepstack in qwen3vl (#26626) · 9d6cff3e

JJJYmmm authored Oct 11, 2025


Signed-off-by: liuye.hj <liuye.hj@alibaba-inc.com>
Signed-off-by: JJJYmmm <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: liuye.hj <liuye.hj@alibaba-inc.com>

9d6cff3e

Add EAGLE-3 Speculative Decoding Support for Qwen3 MoE (#26485) · d2a71530
Rahul Tuli authored Oct 11, 2025
```
Signed-off-by: Rahul Tuli <rtuli@redhat.com>
```
d2a71530

[Refactor]: Use M-RoPE interface directly while defining model class instead... · 727144be

dsinghvi authored Oct 11, 2025


[Refactor]: Use M-RoPE interface directly while defining model class instead of maintaining model specific M-RoPE implementation in mrope.py (#24172)
Signed-off-by: Divyansh Singhvi <divyanshsinghvi@gmail.com>
Signed-off-by: dsinghvi <divyanshsinghvi@gmail.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: wwl2755 <wangwenlong2755@gmail.com>

727144be