Commits · 16f88a8a6c3be2410a17710c6aa9f4a7670bba5e · OpenDAS / vllm_cscc

12 Mar, 2026 1 commit
- moe: 补齐 fill+moe_align 融合开关语义 · 706c031c
  laibao authored Mar 09, 2026
  
  706c031c
09 Feb, 2026 1 commit
- pd separation uses default scheduling and set VLLM_USE_PD_SPLIT=1 · a4bcf959
  zhuwenwen authored Feb 09, 2026
  
  a4bcf959
06 Feb, 2026 1 commit
- 修复awq/w4a16的triton支持，以及fuse_moe的接口对齐，以及awq_moe_marlin推理的相关bug，并解决awq/w4a16的精度问题 · d7db129a
  zhuwenwen authored Feb 06, 2026
  
  d7db129a
04 Feb, 2026 2 commits
- update VLLM_USE_OPT_RESHAPE_AND_CACHE to support bf16 and qwen3-dense · 263f45a4
  zhuwenwen authored Feb 04, 2026
  
  263f45a4
- [perf] add VLLM_USE_FUSED_FILL_RMS_CAT to use lightop for dpsk mtp fill + rms*2 + cat · 5fe03549
  zhuwenwen authored Feb 04, 2026
  
  5fe03549
22 Jan, 2026 1 commit
- [Model] Extend `collect_children` and `no_init_weights` contexts (#32757) · 2b8a38b6
  Cyrus Leung authored Jan 22, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  2b8a38b6
20 Jan, 2026 1 commit
- [Model] Use context managers for encoder- and LM-only mode (#32605) · 4753f3bf
  Cyrus Leung authored Jan 20, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  4753f3bf
16 Jan, 2026 1 commit

解决custom cudagraph模式需要拷贝的问题，需要配合dtk进行使用。 · f1bc9890

zhuwenwen authored Jan 16, 2026

区分pcie和hglink custom allreduce的使用
vllm：export VLLM_CUSTOM_CACHE=1
dtk：export HIP_KERNEL_EVENT_SYSTENFENCE=1

set VLLM_USE_FUSED_RMS_ROPE=1
add SUPPORT_MOE_MARLIN_W16A16 to use moe marlin on bw
support fa kvcache fp8 (todo: add VLLM_USE_QUERY_QUANT to not use q quant)
update moe_align_block_size

f1bc9890

07 Jan, 2026 1 commit
- remove USE_FUSED_RMS_QUANT and USE_FUSED_SILU_MUL_QUANT · 60b37c6b
  zhuwenwen authored Jan 07, 2026
  
  60b37c6b
06 Jan, 2026 1 commit
- 暂不支持awq的scale和zero合并操作 · 2907adfa
  zhuwenwen authored Jan 06, 2026
  
  2907adfa
22 Dec, 2025 1 commit
- [BugFix] skip language model in Encoder (#30242) · 8f8f469b
  dengyunyang authored Dec 22, 2025
```
Signed-off-by: dengyunyang <584797741@qq.com>
```
  8f8f469b
08 Dec, 2025 1 commit

[Model][7/N] Improve all pooling task | Deprecation as_reward_model. Extract... · 9e77ffca

wang.yuqi authored Dec 08, 2025


[Model][7/N] Improve all pooling task | Deprecation as_reward_model. Extract hidden states prefer using new multi-vector retrieval API (#26686)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>

9e77ffca

03 Dec, 2025 2 commits
- set VLLM_USE_LIGHTOP_MOE_ALIGN=0 (for qwen3-30b) · 87503d85
  zhuwenwen authored Dec 03, 2025
  
  87503d85
- add VLLM_USE_OPT_RESHAPE_AND_CACHE、VLLM_USE_FUSE_SILU_AND_MUL and... · 15a55773
  zhuwenwen authored Dec 03, 2025
```
add VLLM_USE_OPT_RESHAPE_AND_CACHE、VLLM_USE_FUSE_SILU_AND_MUL and VLLM_USE_TOPK_RENORM for qwen3-30b
```
  15a55773
02 Dec, 2025 1 commit
- add VLLM_USE_V32_ENCODE to use encoding_dsv32.py · ba7bcccd
  zhuwenwen authored Dec 02, 2025
  
  ba7bcccd
26 Nov, 2025 1 commit
- [Attention] Remove imports from `vllm/attention/__init__.py` (#29342) · 430dd4d9
  Matthew Bonanni authored Nov 26, 2025
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  430dd4d9
22 Nov, 2025 2 commits

[Misc] Fix pre-commit (#29238) · ae668183
Cyrus Leung authored Nov 22, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
ae668183

Fix: Resolve circular import in model_loader/utils.py (#29189) · 6965a392

Nandan Vallamdasu authored Nov 22, 2025

Signed-off-by: nandan2003 <nandan.vallamdasu@outlook.com>
Signed-off-by: Nandan Vallamdasu <nandan.vallamdasu@outlook.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

6965a392

20 Nov, 2025 1 commit
- 修复pd分离开cp引起的校检问题 · acf9f945
  zhuwenwen authored Nov 20, 2025
```
update VLLM_USE_PD_SPLIT=0 (for dspk)and  VLLM_USE_PD_SPLIT=1 (for others)
```
  acf9f945
19 Nov, 2025 1 commit
- Move online quantization to `model.load_weights` (#26327) · da94c7c0
  Jerry Zhang authored Nov 18, 2025
```
Signed-off-by: Jerry Zhang <jerryzh168@gmail.com>
```
  da94c7c0
13 Nov, 2025 2 commits

add VLLM_USE_OPT_ZEROS to replace triton_ (torch.zeros) · e7f2785f
zhuwenwen authored Nov 13, 2025
```
set default_max_num_batched_tokens = 10240
update qwen3_moe of layernorm
off lightop of moe_fused_gate
```
e7f2785f

set VLLM_USE_LIGHTOP=0 for dpsk-v3 · 6a8fd297

zhuwenwen authored Nov 13, 2025

add VLLM_USE_PD_SPLIT to split prefill and decode
replace triton_ of rms and act_and_mul

6a8fd297

23 Oct, 2025 1 commit
- [Chore] Separate out `vllm.utils.platform_utils.py` (#27374) · ca76486a
  Jonathan Chen authored Oct 23, 2025
```
Signed-off-by: Jonathan <chenleejonathan@gmail.com>
```
  ca76486a
18 Oct, 2025 1 commit
- [Chore] Clean up pytorch helper functions in `vllm.utils` (#26908) · 6ac5e06f
  Isotr0py authored Oct 19, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: isotr0py <2037008807@qq.com>
```
  6ac5e06f
12 Oct, 2025 1 commit
- 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
09 Oct, 2025 2 commits

[V0 deprecation] Remove `QKVCrossParallelLinear` implementation (#26475) · d1ddf340
Isotr0py authored Oct 09, 2025
```
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
```
d1ddf340

Separate MLAAttention class from Attention (#25103) · e614ab78

Naveenraj Kamalakannan authored Oct 08, 2025

Signed-off-by: Naveenraj Kamalakannan <therealnaveenkamal@gmail.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

e614ab78

05 Oct, 2025 1 commit
- Convert formatting to use `ruff` instead of `yapf` + `isort` (#26247) · d6953beb
  Harry Mellor authored Oct 05, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  d6953beb
01 Oct, 2025 1 commit
- Support RL online quantization with torchao (#23014) · c3124680
  Jerry Zhang authored Oct 01, 2025
```
Signed-off-by: Jerry Zhang <jerryzh168@gmail.com>
```
  c3124680
26 Sep, 2025 1 commit
- update VLLM_USE_LIGHTOP and VLLM_USE_OPT_CAT · f8cf43ae
  zhuwenwen authored Sep 26, 2025
  
  f8cf43ae
25 Sep, 2025 2 commits
- [Optimization] Use a cheaper cache key in `get_model_architecture` (#25682) · 89fa54e6
  Cyrus Leung authored Sep 26, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  89fa54e6
- update VLLM_USE_LIGHTOP and VLLM_USE_OPT_CAT · f70d0ec6
  zhuwenwen authored Sep 25, 2025
  
  f70d0ec6
20 Sep, 2025 1 commit
- [Optimization] Avoid repeated model architecture conversion for pooling models (#25261) · c60e6137
  Cyrus Leung authored Sep 20, 2025
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  c60e6137
19 Sep, 2025 1 commit
- Move `ModelConfig` from `config/__init__.py` to `config/model.py` (#25252) · aed16879
  Harry Mellor authored Sep 19, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  aed16879
13 Sep, 2025 1 commit
- Enable conversion of multimodal models to pooling tasks (#24451) · bbb70036
  Maximilien de Bayser authored Sep 12, 2025
```
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
```
  bbb70036
12 Sep, 2025 1 commit
- Enable conversion of multimodal models to pooling tasks (#24451) · e090b7b4
  Maximilien de Bayser authored Sep 12, 2025
```
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
```
  e090b7b4
08 Sep, 2025 1 commit
- [Model] Remove quantized mixtral (#24437) · 8d7f39b4
  Jee Jee Li authored Sep 09, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  8d7f39b4
28 Jul, 2025 1 commit

[Deprecation][2/N] Replace `--task` with `--runner` and `--convert` (#21470) · 86ae693f

Cyrus Leung authored Jul 28, 2025

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

86ae693f

24 Jul, 2025 1 commit
- [Model] Officially support Emu3 with Transformers backend (#21319) · 13abd0ea
  Harry Mellor authored Jul 24, 2025
```
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  13abd0ea
23 Jul, 2025 1 commit
- add clear messages for deprecated models (#21424) · 2f5c14de
  youkaichao authored Jul 23, 2025
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  2f5c14de