Commit Graph

95 Commits

Author SHA1 Message Date
Brayden Zhong
006ead9dcb [FA][Test] Fix Sparse FA test (#6306) 2025-05-26 01:27:48 -07:00
Yuan Luo
121f92c583 Add main for merge state tests (#6492)
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
2025-05-21 21:56:25 -07:00
HandH1998
4d643f6c7a [1/2] Support Qserve (#6457)
Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
2025-05-21 19:48:59 -07:00
Elfie Guo
6fc9357503 [2/2] Add python wrapper for CUTLASS FP8 Blockscale MoE Kernel. (#5694) 2025-05-16 13:14:07 -07:00
Lianmin Zheng
e8e18dcdcc Revert "fix some typos" (#6244) 2025-05-12 12:53:26 -07:00
applesaucethebun
d738ab52f8 fix some typos (#6209)
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
2025-05-13 01:42:38 +08:00
applesaucethebun
2ce8793519 Add typo checker in pre-commit (#6179)
Co-authored-by: Brayden Zhong <b8zhong@uwaterloo.ca>
2025-05-11 12:55:00 +08:00
Trevor Morris
0ab3f437ab Cutlass MLA: Disable split kv due to https://github.com/NVIDIA/cutlass/issues/2274 (#6101) 2025-05-08 18:44:30 -07:00
PGFLMG
08acdb5c3d [Feat] Scale up fa3 kernel to sm8x arch (#5912)
Co-authored-by: zhyncs <me@zhyncs.com>
2025-04-30 13:59:36 -07:00
PGFLMG
ee71ed8a41 [Feat] QWen-1M context support[1/2]: Update block sparse attention backend utils kernel (#5847)
Co-authored-by: sighingnow <sighingnow@gmail.com>
2025-04-28 11:03:17 -07:00
Yineng Zhang
15fabcc07f fix sgl-kernel unit tests (#5666) 2025-04-23 01:18:30 -07:00
Elfie Guo
e62c49557d [1/2] Add FP8 Blockscale MoE CUTLASS kernel for Blackwell (#5281) 2025-04-22 22:28:20 -07:00
Xiaoyu Zhang
8e09b37077 Sgl kernel fused_moe_gate support n_shared_experts (#5440) 2025-04-17 23:05:15 -07:00
PGFLMG
c08a717c77 [Feat] Update sgl-kernel flashinfer to latest main version (#5500)
Co-authored-by: zhyncs <me@zhyncs.com>
2025-04-17 12:43:23 -07:00
Baizhou Zhang
81c891111f Add test for flash_attn_varlen_func kernel (#5484) 2025-04-17 01:42:56 -07:00
Trevor Morris
e8f62b20ca BLackwell cutlass mla: Add check for bad page size/block num combinations (#5431) 2025-04-15 14:07:42 -07:00
DefTruth
388e15c0db kernel: support slightly faster merge_state_v2 cuda kernel (#5381) 2025-04-14 21:28:23 -07:00
Yineng Zhang
b62e7e99b8 feat: adapt merge_state (#5337) 2025-04-12 21:14:04 -07:00
PGFLMG
4879e50c6d [Feat] Add sparse attn to sgl-kernel (#5327) 2025-04-12 11:36:36 -07:00
Zhaoyi Li
3c9740d200 update variable naming and comments for rocm (#5299) 2025-04-11 23:15:05 -07:00
Trevor Morris
f65b8d5c89 Blackwell Cutlass MLA kernel (#5142) 2025-04-11 22:16:51 -07:00
Yineng Zhang
136b8e6afb fix: remove cublas_grouped_gemm (#5307) 2025-04-11 16:22:37 -07:00
Yineng Zhang
c1dd773c19 fix: use fa3 unit test on hopper only (#5304) 2025-04-11 15:10:49 -07:00
PGFLMG
ed01b4515e [Misc] Clean sgl-kernel test (#5216) 2025-04-10 11:28:41 -07:00
Xiaoyu Zhang
f730362ee2 reduce moe_align_block_size_kernel small batch mode overhead (#5086) 2025-04-09 17:59:35 -07:00
Yi Zhang
ebf495f013 sgl-kernel use cutlass latest version for fp8 blockwise gemm (#5207) 2025-04-09 11:47:04 -07:00
yinfan98
d2e507df3c [Misc] clean up vllm in sgl-kernel test (#5189) 2025-04-09 01:22:13 -07:00
yinfan98
9798e72baa [Misc] Use pytest.mark.skipif in sgl-kernel test (#5137) 2025-04-07 21:35:14 -07:00
Yi Zhang
bcbbf519f9 sgl-kernel transfer custom allreduce from trt kernel to vllm kernel (#5079) 2025-04-05 14:23:20 -07:00
Xiaoyu Zhang
924ca7c92c Add DeepSeek V3/R1 shared experts fusion (#4918) 2025-04-04 01:59:29 -07:00
yinfan98
b8b6008f47 [Fix] fix fa3 build at cu118 (#5036) 2025-04-03 11:52:35 -07:00
Xiaoyu Zhang
2c8fd99363 [sgl-kernel] per token group quant support COLUMN MAJOR (#4817) 2025-04-02 18:29:59 -07:00
yinfan98
37c66ec856 [feat] add fa3 in sgl-kernel (#4902)
Co-authored-by: Sleepcoo <Sleepcoo@gmail.com>
2025-03-30 12:57:10 -07:00
Adarsh Shirawalmath
9fccda3111 [Feature] use pytest for sgl-kernel (#4896) 2025-03-30 10:36:52 -07:00
Yi Zhang
5ec5eaf760 fix allreduce test (#4909) 2025-03-29 23:16:53 -07:00
Qingquan Song
45dcfc2e76 Add deepseek style fused moe group gate selection kernel (#4530) 2025-03-29 11:51:45 -07:00
yinfan98
ddf8981d91 Delete test_deep_gemm.py (#4891) 2025-03-29 10:46:11 -07:00
Trevor Morris
e9f8e42318 Support FP4 gemm (1/2) (#3899) 2025-03-24 19:50:23 -07:00
Chunan Zeng
65c24c28f9 [Quant Kernel] refactored per token group quant fp8 to support int8 up-to 2x faster (#4396) 2025-03-23 23:44:17 -07:00
AniZpZ
321ab756bc [1/3] fix dsv3 awq issue (#4556)
Co-authored-by: leoneo <1320612015@qq.com>
2025-03-22 01:07:17 -07:00
Wenbo Yang
75b656488a Support serving DeepSeek-R1-Channel-INT8 with 32 L40S. (#4418) 2025-03-17 00:03:43 -07:00
Yineng Zhang
9971dc2283 Revert "feat: Add FlashMLA submodule (#4449)" (#4470) 2025-03-16 01:30:05 -07:00
Ying Sheng
52a34d7448 Add greedy verification kernel (#4383) 2025-03-16 00:58:26 -07:00
JieXin Liang
1a3fa75f2f [Fix] use torch.cat instead of torch.concat to prevent entering the Autograd backends. (#4466) 2025-03-16 00:02:47 -07:00
Shi Shuai
81f431eded feat: Add FlashMLA submodule (#4449)
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
2025-03-15 23:30:25 -07:00
Qingquan Song
61e4433caf Add moe topk softmax templated from vllm (#4302) 2025-03-14 12:03:33 -07:00
Yineng Zhang
2937387a50 fix accuracy issue (#4376) 2025-03-13 02:06:22 -07:00
Qingquan Song
4068e01292 Fix per token fp8 quant precision (#4362) 2025-03-12 21:19:05 -07:00
Rex
07f944631e Add awq dequantize kernel to sgl with 1x to 3x speedup (#4104) 2025-03-12 00:10:02 -07:00
Xiaoyu Zhang
23308a9032 fix per_token_group_quant_fp8 illegal memory when num_groups % 16 != 0 (#4231) 2025-03-10 01:42:58 -07:00