Commits · f8f9244a61544fd90800d94859ed5609798e100d · zhaoyu6 / sglang

19 Mar, 2025 2 commits
- avoid cudaStreamSynchronize in DeepSeekV2AttentionMLA (#4577) · df7014a8
  strgrb authored Mar 20, 2025
```
Co-authored-by: Zhang Kaihong <zhangkaihong.zkh@alibaba-inc.com>
```
  df7014a8
- [Feature] Integrate DeepEP into SGLang (#4232) · f44db16c
  Jinyan Chen authored Mar 19, 2025
```
Co-authored-by: Cheng Wan <cwan39@gatech.edu>
Co-authored-by: Xuting Zhou <xutingz@nvidia.com>
```
  f44db16c
18 Mar, 2025 2 commits
- Reduce computation and communication in DP attention (#4521) · 3196999f
  Cheng Wan authored Mar 18, 2025
  
  3196999f
- cleanup deps 3/n (#4541) · c16b33cc
  Yineng Zhang authored Mar 18, 2025
  
  c16b33cc
17 Mar, 2025 1 commit

[Feature] Support Deepseek-VL2 (#2798) · d6d21640

萝卜菜 authored Mar 17, 2025


Co-authored-by: Edenzzzz <wtan45@wisc.edu>
Co-authored-by: Chayenne <zhaochen20@outlook.com>
Co-authored-by: Yi Zhang <1109276519@qq.com>

d6d21640

16 Mar, 2025 1 commit
- cleanup deps 2/n (#4464) · 65b7c9b7
  Yineng Zhang authored Mar 15, 2025
  
  65b7c9b7
13 Mar, 2025 3 commits
- Improve DP attention (#4390) · 8e66fbec
  Lianmin Zheng authored Mar 13, 2025
```
Co-authored-by: dhou-xai <dhou@x.ai>
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
```
  8e66fbec
- Revert "[XPU][CPU] Enable the native path of DeepSeek" (#4367) · 45de8971
  Lianmin Zheng authored Mar 12, 2025
  
  45de8971
- [XPU][CPU] Enable the native path of DeepSeek (#4086) · 71046fcd
  Meng, Hengyu authored Mar 13, 2025
```
Co-authored-by: Zhang, Liangang <liangang.zhang@intel.com>
```
  71046fcd
12 Mar, 2025 1 commit
- unify is_cuda and is_hip (#4321) · d1da58e2
  Yineng Zhang authored Mar 11, 2025
  
  d1da58e2
10 Mar, 2025 1 commit
- [Bug fixed] fixed the crash when enable the dp-attention on the single card (#3958) · 4455b26e
  DavidChan authored Mar 10, 2025
  
  4455b26e
09 Mar, 2025 1 commit
- Support nextn for flashinfer mla attention backend (#4218) · 9fb48f95
  Baizhou Zhang authored Mar 09, 2025
  
  9fb48f95
07 Mar, 2025 1 commit

[Feature] DeepSeek V3/R1 INT8 Quantization (channel-wise) (#3888) · c7f25446

HandH1998 authored Mar 07, 2025


Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
Co-authored-by: b0urnee <2769086541@qq.com>

c7f25446

05 Mar, 2025 2 commits
- [QUANT] Add GPTQModel Dynamic Quantization + `lm_head` Quantization (#3790) · 56a724eb
  Qubitium-ModelCloud authored Mar 05, 2025
```
Signed-off-by: ZX-ModelCloud <zx@modelcloud.ai>
Co-authored-by: ZX-ModelCloud <zx@modelcloud.ai>
```
  56a724eb
- [Minor] more code cleanup (#4077) · e074d84e
  Lianmin Zheng authored Mar 04, 2025
  
  e074d84e
03 Mar, 2025 1 commit
- Share target model embed and head weights for nextn (#4033) · 9fafa62d
  Ke Bao authored Mar 04, 2025
  
  9fafa62d
01 Mar, 2025 1 commit

[Feature]Support ragged prefill in flashinfer mla backend (#3967) · 90a4b7d9

Baizhou Zhang authored Feb 28, 2025


Co-authored-by: Yineng Zhang <me@zhyncs.com>
Co-authored-by: pankajroark <pankajroark@users.noreply.github.com>

90a4b7d9

25 Feb, 2025 2 commits
- Fix allgather ops inside cuda graphs (#3709) · 127998cc
  Nicolas Castet authored Feb 25, 2025
  
  127998cc
- [ROCm] Enable Fused MLA Triton kernel for DeepSeekV3 (#3237) · 6ce9dbe8
  Chaitanya Sri Krishna Lolla authored Feb 25, 2025
```
Co-authored-by: HAI <hixiao@gmail.com>
```
  6ce9dbe8
24 Feb, 2025 2 commits
- Feature DeepSeek V3/R1 INT8 Quantization (block-wise) (#3730) · 1a6e9757
  laixin authored Feb 24, 2025
```
Co-authored-by: HandH1998 <1335248067@qq.com>
```
  1a6e9757
- Refactor flashinfer logic for deepseek v3 and fix accuracy bug (#3785) · b1100846
  Baizhou Zhang authored Feb 24, 2025
  
  b1100846
17 Feb, 2025 1 commit
- feat: support flashinfer mla with prefix cache (#3643) · 714f3e63
  Yineng Zhang authored Feb 18, 2025
  
  714f3e63
14 Feb, 2025 2 commits
- Support NextN (MTP) speculative decoding for DeepSeek-V3/R1 (#3582) · 862dd76c
  Ke Bao authored Feb 15, 2025
  
  862dd76c
- feat: support flashinfer mla attention for deepseek v3 (#3550) · 70f894b8
  Yineng Zhang authored Feb 14, 2025
  
  70f894b8
12 Feb, 2025 1 commit
- Fix deepseek awq v3 (#3450) · 8616357a
  Liangsheng Yin authored Feb 12, 2025
  
  8616357a
27 Jan, 2025 1 commit
- feat: use sgl-kernel 0.0.3 in sglang (#3179) · 2f79f588
  Yineng Zhang authored Jan 27, 2025
  
  2f79f588
19 Jan, 2025 1 commit
- fix deepseek v2 with cpu device (#2975) · 5a176c92
  Yineng Zhang authored Jan 19, 2025
  
  5a176c92
18 Jan, 2025 1 commit
- feat: remove vllm get_rope (#2964) · 2add697d
  Yineng Zhang authored Jan 18, 2025
  
  2add697d
17 Jan, 2025 2 commits
- feat: remove vllm distributed (#2907) · 5dc54f1a
  Yineng Zhang authored Jan 17, 2025
```
Co-authored-by: Zhangyi <1109276519@qq.com>
```
  5dc54f1a
- Enable CPU device on SGLang (#2806) · 63051738
  Chunyuan WU authored Jan 17, 2025
  
  63051738
16 Jan, 2025 1 commit
- Support multi-node DP attention (#2925) · 8b6ce52e
  Lianmin Zheng authored Jan 16, 2025
```
Co-authored-by: dhou-xai <dhou@x.ai>
```
  8b6ce52e
30 Dec, 2024 1 commit
- AMD DeepSeek_V3 FP8 Numerical fix (#2667) · c5210dfa
  HAI authored Dec 30, 2024
  
  c5210dfa
25 Dec, 2024 1 commit
- Refactor MoE (#2575) · 53aed988
  HandH1998 authored Dec 26, 2024
```
Co-authored-by: zhyncs <me@zhyncs.com>
```
  53aed988
23 Dec, 2024 1 commit
- Reorg moe code (#2563) · e835a500
  Ke Bao authored Dec 24, 2024
  
  e835a500
06 Dec, 2024 1 commit
- Resubmit MoE-EP (#2371) · 3d32e4a3
  xiaobochen authored Dec 06, 2024
  
  3d32e4a3
05 Dec, 2024 1 commit
- Fix AWQ with enable MLA (#2364) · 4a63c181
  Ke Bao authored Dec 06, 2024
  
  4a63c181
04 Dec, 2024 1 commit
- MLA prefill w/o weight absorption (#2349) · ec52464d
  Ke Bao authored Dec 05, 2024
  
  ec52464d
02 Dec, 2024 1 commit
- Update model_loader deps and qqq quantization deps (#2220) (#2318) · 85e1a6f3
  Yineng Zhang authored Dec 02, 2024
```
Co-authored-by: HandH1998 <1335248067@qq.com>
```
  85e1a6f3
01 Dec, 2024 2 commits
- Revert "Revert "[FEAT] Support GGUF format"" (#2287) · 4936be8a
  Lianmin Zheng authored Nov 30, 2024
  
  4936be8a
- Revert "[FEAT] Support GGUF format" (#2285) · 7e4c6dd8
  Lianmin Zheng authored Nov 30, 2024
  
  7e4c6dd8