Commits · fe3398fab2b11f92b4ada209a85558710c446a36 · OpenDAS / vllm_cscc

29 Nov, 2025 1 commit
- [Chore] Enable passing `tokenizer=None` into MM processor (#29724) · fe3398fa
  Cyrus Leung authored Nov 29, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  fe3398fa
28 Nov, 2025 3 commits

Remove upstream fa checks (#29471) · 460d8bbf

Mingyuan Ma authored Nov 28, 2025


Signed-off-by: mingyuanm <mingyuanm@nvidia.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>

460d8bbf

Revert "[CPU]Update CPU PyTorch to 2.9.0 (#29589)" (#29647) · b34e8775
Cyrus Leung authored Nov 28, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
b34e8775

[Multimodal][Speculative Decoding]Eagle3 mm support, enablement on qwen3vl (#29594) · 37b15e97

EanWang211123 authored Nov 28, 2025


Signed-off-by: Tsai, Louie <louie.tsai@intel.com>
Signed-off-by: EanWang211123 <wangyiheng@sangfor.com.cn>
Co-authored-by: Louie Tsai <louie.tsai@intel.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>

37b15e97

24 Nov, 2025 1 commit
- [Core] Deprecate `xformers` (#29262) · 0ff70821
  Roger Wang authored Nov 23, 2025
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  0ff70821
20 Nov, 2025 1 commit
- [Model][QwenVL] Replace `torch.repeat_interleave` with faster `np.repeat` (#28964) · a9705a29
  Lukas Geiger authored Nov 20, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  a9705a29
19 Nov, 2025 2 commits
- Update `rope_scaling` to `rope_parameters` in preparation for Transformers v5 (#28542) · a8b70304
  Harry Mellor authored Nov 19, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  a8b70304
- [Model][QwenVL] Simplify cos/sin rotary embedding indexing (#28962) · 3d4e7d34
  Lukas Geiger authored Nov 19, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  3d4e7d34
18 Nov, 2025 1 commit
- [Model][Perf] Use cos and sin cache in QwenVL (#28798) · b9489f51
  Canlin Guo authored Nov 18, 2025
```
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
```
  b9489f51
15 Nov, 2025 2 commits
- [Model][Qwen3VL] Cache positional embedding indices (#28475) · 07cadab2
  Lukas Geiger authored Nov 15, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
```
  07cadab2
- [Model][Qwen3VL] Use `mm_position` to compute mrope positions (#28730) · f05d474c
  Lukas Geiger authored Nov 15, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  f05d474c
14 Nov, 2025 2 commits

[Bugfix] resolve Qwen3-VL GPTQModel quantized model loading failure (#28663) · cec275ef

GuanH authored Nov 15, 2025


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

cec275ef

[Model][MM] Extract conv layer as CustomOp (#28455) · 41b92f7d

Shanshan Shen authored Nov 14, 2025


Signed-off-by: shen-shanshan <467638484@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

41b92f7d

13 Nov, 2025 1 commit
- Rename clashing method names for vLLM model protocol (#27583) · 97d1c993
  Harry Mellor authored Nov 13, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  97d1c993
12 Nov, 2025 1 commit
- [Model][Qwen3VL] Simplify `get_mrope_input_positions` using numpy (#28302) · cbb799e3
  Lukas Geiger authored Nov 12, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  cbb799e3
11 Nov, 2025 4 commits
- [LoRA][1/N]Remove LoRA extra vocab (#28382) · 9d1c4747
  Jee Jee Li authored Nov 12, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  9d1c4747
- [Model] Pass `mm_features` directly into `get_mrope_input_positions` (#28399) · afffd3cc
  Cyrus Leung authored Nov 11, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  afffd3cc
- [Attention] Refactor CUDA attention backend selection logic (#24794) · b30dfa03
  Matthew Bonanni authored Nov 11, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
```
  b30dfa03
- [Model][Qwen3VL] Slighly speedup `fast_pos_embed_interpolate` (#28434) · 9973e6e0
  Lukas Geiger authored Nov 11, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  9973e6e0
10 Nov, 2025 1 commit
- [V0 Deprecation] Remove unused `context_len` and `seq_len` from M-RoPE (#28395) · d0e186c1
  Cyrus Leung authored Nov 11, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d0e186c1
07 Nov, 2025 1 commit
- [Core][MM] Add mechanism to configure multimodal fields which should stay on CPU (#28168) · e0919f33
  Lukas Geiger authored Nov 07, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  e0919f33
29 Oct, 2025 1 commit

[Model] Fix Qwen3VL and Qwen3Omni after torch.compile changes (#27705) · 0d8161b0

Lukas Geiger authored Oct 29, 2025


Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>

0d8161b0

27 Oct, 2025 1 commit
- [Model] Use merge_by_field_config for MM models (Qwen series) (#27546) · cbd5e07a
  Cyrus Leung authored Oct 27, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  cbd5e07a
25 Oct, 2025 1 commit

[CI/Build] Refactor processing tests (#27470) · 66a168a1

Cyrus Leung authored Oct 26, 2025


Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

66a168a1

24 Oct, 2025 1 commit
- [MM][Bugfix] Replace `PatchEmbed`'s conv3d to linear layer (#27418) · 42efe609
  Isotr0py authored Oct 24, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Roger Wang <hey@rogerw.io>
```
  42efe609
22 Oct, 2025 1 commit
- [Bugfix] Make `get_mrope_input_positions` instance methods (#27342) · 14e2f123
  Cyrus Leung authored Oct 22, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  14e2f123
21 Oct, 2025 1 commit
- [MM][Core] Decouple ViT backend from LM backend (#27061) · c3a2c6ac
  Roger Wang authored Oct 21, 2025
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  c3a2c6ac
19 Oct, 2025 1 commit
- [Misc] Move utils to avoid conflicts with stdlib, and move tests (#27169) · d31f7844
  Cyrus Leung authored Oct 19, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d31f7844
17 Oct, 2025 1 commit
- [bugfix] Qwen3-VL fix video incorrect timestamp calculations while do_sample_frames=True (#27104) · 4c91a28e
  燃 authored Oct 18, 2025
```
Co-authored-by: 松灵 <wpf272043@alibaba-inc.com>
```
  4c91a28e
16 Oct, 2025 2 commits
- [Model] Fix Qwen3VL mm mapping (#27027) · 9f4e3090
  Jee Jee Li authored Oct 17, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  9f4e3090
- [Chore] Separate out `vllm.utils.collections` (#26990) · d2740faf
  Cyrus Leung authored Oct 16, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d2740faf
14 Oct, 2025 1 commit
- [Bugfix] Standardize merging multimodal embeddings (#26771) · d2f816d6
  Cyrus Leung authored Oct 14, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  d2f816d6
12 Oct, 2025 2 commits
- [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
- 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
11 Oct, 2025 2 commits

[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

[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

10 Oct, 2025 1 commit

[Model][Qwen3VL] Compute `cu_seqlens` on CPU to remove (#26496) · b2155ed3

Lukas Geiger authored Oct 10, 2025


Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

b2155ed3

09 Oct, 2025 2 commits
- [Models][Qwen] Replace `pad` with `cat` for better performance (#26486) · 2c1c7dfb
  Lukas Geiger authored Oct 09, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  2c1c7dfb
- [Models][Qwen3VL] Optimise `_validate_and_reshape_mm_tensor` (#26426) · 0426e3c5
  Lukas Geiger authored Oct 09, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  0426e3c5
08 Oct, 2025 1 commit
- [Models] Improve iteration over layers (#26425) · 93f2c0aa
  Lukas Geiger authored Oct 08, 2025
```
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
```
  93f2c0aa