Commits · f98e88b9fbbb59ad700892da765bc49bda34c59b · change / sglang

19 Jul, 2025 1 commit
- chore: bump sgl-kernel v0.2.6 (#8165) · f98e88b9
  Yineng Zhang authored Jul 19, 2025
  
  f98e88b9
16 Jul, 2025 1 commit
- [1/n] chore: decouple quantization implementation from vLLM dependency (#7992) · c28ad199
  Peng Zhang authored Jul 17, 2025
  
  c28ad199
14 Jul, 2025 1 commit
- [Feature] CUDA Green Context Support (#7649) · 1ebec1a8
  ykcombat authored Jul 15, 2025
  
  1ebec1a8
12 Jul, 2025 1 commit
- chore: bump sgl-kernel v0.2.5 (#7964) · 8f3173d0
  Yineng Zhang authored Jul 11, 2025
  
  8f3173d0
05 Jul, 2025 4 commits
- chore: bump sgl-kernel v0.2.4 (#7800) · f200af0d
  Yineng Zhang authored Jul 05, 2025
  
  f200af0d
- Add treemask mode to build_eagle_tree & release sgl-kernel 0.2.3 (#7756) · 5589b750
  Lianmin Zheng authored Jul 05, 2025
```
Co-authored-by: Pranjal Shankhdhar <pranjal.ssh@gmail.com>
```
  5589b750
- chore: bump sgl-kernel v0.2.3 (#7784) · 4fece12b
  Yineng Zhang authored Jul 05, 2025
  
  4fece12b
- [1/n]: add cutlass W4A8 moe kernel for hopper architecture (#7772) · da3890e8
  SijiaYang authored Jul 05, 2025
```
Signed-off-by: yangsijia.614 <yangsijia.614@bytedance.com>
Co-authored-by: yicwang <yichen.wang@bytedance.com>
```
  da3890e8
03 Jul, 2025 2 commits
- chore: bump sgl-kernel 0.2.2 (#7755) · aca1101a
  Yineng Zhang authored Jul 03, 2025
  
  aca1101a
- [optimize] fuse renormalize into moe_topk_softmax (#7744) · 2998c4bd
  Yi Zhang authored Jul 04, 2025
```
Co-authored-by: ispobock <ispobaoke@gmail.com>
```
  2998c4bd
02 Jul, 2025 1 commit

[1/n] apply wna16marlin kernel in moe weight only quantization (#7683) · 8e03b641

AniZpZ authored Jul 02, 2025


Co-authored-by: 晟海 <huangtingwei.htw@antgroup.com>
Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: HandH1998 <1335248067@qq.com>
Co-authored-by: 弋云 <yiyun.wyt@antgroup.com>
Co-authored-by: walker-ai <2398833647@qq.com>

8e03b641

01 Jul, 2025 1 commit
- chore: bump sgl-kernel v0.2.1 (#7675) · 637bfee4
  Yineng Zhang authored Jun 30, 2025
  
  637bfee4
30 Jun, 2025 1 commit
- Add dsv3 router gemm kernel (#7627) · 7248272c
  Baizhou Zhang authored Jun 29, 2025
  
  7248272c
29 Jun, 2025 1 commit
- Add dsv3 fused a gemm to sgl-kernel (#7630) · 04b35190
  Ke Bao authored Jun 29, 2025
  
  04b35190
25 Jun, 2025 1 commit
- Fuse sorted_token_ids padding to moe_align_block_size kernel (#7437) · 57ab7769
  Ke Bao authored Jun 25, 2025
  
  57ab7769
24 Jun, 2025 1 commit
- chore: bump sgl-kernel v0.2.0 (#7490) · e846d95e
  Yineng Zhang authored Jun 23, 2025
  
  e846d95e
23 Jun, 2025 2 commits
- kvcache io kernels and test case (#7382) · 34c3f9b2
  Zhiqiang Xie authored Jun 23, 2025
  
  34c3f9b2
- Fix a bug in BatchTokenIDOut & Misc style and dependency updates (#7457) · 55e03b10
  Lianmin Zheng authored Jun 23, 2025
  
  55e03b10
16 Jun, 2025 2 commits
- Release sgl-kernel 0.1.9 (#7232) · 8321f8e4
  Lianmin Zheng authored Jun 16, 2025
  
  8321f8e4
- Fix sampling for speculative decoding & simplify kernels (#7207) · cfceb83d
  Lianmin Zheng authored Jun 16, 2025
  
  cfceb83d
15 Jun, 2025 1 commit
- chore: bump v0.1.8.post2 (#7189) · 44733203
  Yineng Zhang authored Jun 14, 2025
  
  44733203
14 Jun, 2025 1 commit
- [fix] fix cutlass_mla_backend with cuda_graph and add sm_scale for sgl-kernel cutlass_mla (#7184) · ab1a4fa5
  JieXin Liang authored Jun 15, 2025
  
  ab1a4fa5
13 Jun, 2025 3 commits
- chore: bump v0.1.8.post1 (#7152) · 8ab7d93c
  Yineng Zhang authored Jun 13, 2025
  
  8ab7d93c
- Support new DeepGEMM format in per token group quant (#7146) · 5c66c442
  fzyzcjy authored Jun 13, 2025
  
  5c66c442
- Remove 200us slow concat kernel (part 1: kernel) (#7145) · aa46ed34
  fzyzcjy authored Jun 13, 2025
  
  aa46ed34
12 Jun, 2025 2 commits
- feat: update blackwell setup (#7119) · 7046e0fa
  Yineng Zhang authored Jun 12, 2025
  
  7046e0fa
- [sgl-kernel] Add cuda kernel for moe_ep_silu_and_mul (#6919) · 84727a51
  Yuan Luo authored Jun 12, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  84727a51
09 Jun, 2025 1 commit
- [perf][sgl-kernel] extend cutlass_mla_decode to support num_head < 128 (#6929) · 18efb5e8
  JieXin Liang authored Jun 09, 2025
  
  18efb5e8
08 Jun, 2025 1 commit
- chore: bump sgl-kernel v0.1.7 (#6963) · 6c0a4828
  Yineng Zhang authored Jun 08, 2025
  
  6c0a4828
07 Jun, 2025 3 commits
- chore: bump sgl-kernel v0.1.6.post1 (#6955) · 8db3ac55
  Yineng Zhang authored Jun 07, 2025
  
  8db3ac55
- Add a CUDA kernel for fusing mapping and weighted sum for MoE. (#6916) · 3e56f557
  Elfie Guo authored Jun 07, 2025
```
Co-authored-by: Elfie Guo <elfiegxf@gmail.com>
```
  3e56f557
- chore: bump sgl-kernel v0.1.6 (#6943) · d664ca18
  Yineng Zhang authored Jun 07, 2025
  
  d664ca18
05 Jun, 2025 3 commits
- [CUTLASS-FP4-MOE] Introduce CutlassMoEParams class for easy initialization of... · 0df6765c
  Pavani Majety authored Jun 05, 2025
```
[CUTLASS-FP4-MOE]  Introduce CutlassMoEParams class for easy initialization of Cutlass Grouped Gems Metadata (#6887)
Signed-off-by: Pavani Majety <pmajety@nvidia.com>
```
  0df6765c
- [EP] Add cuda kernel for moe_ep_post_reorder (#6837) · 43baba64
  Yuan Luo authored Jun 05, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  43baba64
- support 1 shot allreduce in 1-node and 2-node using mscclpp (#6277) · 8e3797be
  zyksir authored Jun 05, 2025
  
  8e3797be
04 Jun, 2025 2 commits
- Set `num_fused_shared_experts` as `num_shared_experts` when shared_experts... · 81964328
  Cheng Wan authored Jun 04, 2025
```
Set `num_fused_shared_experts` as `num_shared_experts` when shared_experts fusion is not disabled (#6736)
```
  81964328
- [Refactor] Rename `n_share_experts_fusion` as `num_fused_shared_experts` (#6735) · 8a548052
  Cheng Wan authored Jun 03, 2025
  
  8a548052
02 Jun, 2025 2 commits
- [1/2] Add Kernel support for Cutlass based Fused FP4 MoE (#6093) · eb38c7d1
  Pavani Majety authored Jun 02, 2025
```
Signed-off-by: Pavani Majety <pmajety@nvidia.com>
```
  eb38c7d1
- [EP] Add cuda kernel for moe_ep_pre_reorder (#6699) · 55444ed6
  Yuan Luo authored Jun 02, 2025
```
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
```
  55444ed6
01 Jun, 2025 1 commit
- misc: cache is_hopper_arch (#6799) · c429919d
  Wenxuan Tan authored Jun 01, 2025
  
  c429919d