Commit Graph

2775 Commits

Author SHA1 Message Date
Yineng Zhang
c1dd773c19 fix: use fa3 unit test on hopper only (#5304) 2025-04-11 15:10:49 -07:00
Yineng Zhang
6f8593799b feat: add blackwell workflow (#5303) 2025-04-11 13:42:00 -07:00
Yineng Zhang
f774a0d275 feat: add blackwell Dockerfile (#5302) 2025-04-11 13:08:53 -07:00
Xiaoyu Zhang
60bcbf2a35 remove moe_align_block_size torch.zeros in small batch/expert mode (#5298) 2025-04-11 12:13:55 -07:00
Adarsh Shirawalmath
a0a9f6d64f [Docs] Remove the older supported docs section (#5301) 2025-04-11 11:30:18 -07:00
Yineng Zhang
80aa8ca84e fix: update update_wheel_index for cu128 (#5300) 2025-04-11 09:31:03 -07:00
Adarsh Shirawalmath
4aa6bab0b0 [Docs] Supported Model Docs - Major restructuring (#5290)
Co-authored-by: zhaochenyang20 <zhaochen20@outlook.com>
2025-04-11 09:17:47 -07:00
Yusong Gao
c35dcfdb30 [PD] fix: skip warmup request in disaggregation mode to prevent crash on timeout (#5292) 2025-04-11 23:03:07 +08:00
Yineng Zhang
c163bf4ff1 chore: bump sgl-kernel v0.0.8.post1 (#5289) 2025-04-11 02:11:53 -07:00
Yineng Zhang
5598634326 chore: relax the torch version restriction for sgl-kernel compilation (#5288) 2025-04-11 02:05:53 -07:00
Yineng Zhang
b75275b6f2 feat: add cu128 identifier for sgl-kernel (#5287) 2025-04-11 01:58:46 -07:00
Yineng Zhang
7074e9ca20 fix: enable fp4 compilation on cu128 (#5286) 2025-04-11 01:43:44 -07:00
Michael Yao
fc14cca088 Fix a 404 link in send_request.ipynb (#5280)
Signed-off-by: windsonsea <haifeng.yao@daocloud.io>
2025-04-11 01:38:45 -07:00
XinyuanTong
e7beff8a13 fix: examples for token_in_token_out_vlm (#5193) 2025-04-11 01:38:23 -07:00
mlmz
4d2e305149 doc: nested loop code for offline engine (#5244) 2025-04-11 01:36:30 -07:00
Mick
e53a0b3d5b [fix] fix mrope positions not picked up (#5265) 2025-04-11 01:29:45 -07:00
Cheng Wan
038bc5d521 Support --enable-llama4-multimodal (#5254) 2025-04-11 01:24:14 -07:00
Chang Su
aee62d744b Optimize GPU memory usage in FlashAttentionBackend's strided indexing (#5262)
Co-authored-by: ch-wan <cwan39@gatech.edu>
2025-04-11 00:34:17 -07:00
fzyzcjy
cd7e32e2cb Optimize attention in llama4 (#5127) 2025-04-11 00:32:41 -07:00
HAI
8879944800 ROCm/AITER CK_MoE: update 2-stage kernels & support both Activations (#5228) 2025-04-10 18:19:57 -07:00
Richard Zou
a879811c4b Fix torch.compile cacheing (#5259)
Co-authored-by: zhyncs <me@zhyncs.com>
2025-04-10 18:08:45 -07:00
Elfie Guo
a222945df2 Update Makefile / build script to avoid installing incompatible torch dependency (#5245) 2025-04-10 22:21:02 +00:00
PGFLMG
ed01b4515e [Misc] Clean sgl-kernel test (#5216) 2025-04-10 11:28:41 -07:00
HAI
d050df368c ROCm sgl-kernel: compatible to later torch (#5167) 2025-04-10 09:18:36 -07:00
Richard Zou
76f44c2a8d Fix deepseek-v3 with torch.compile in PyTorch 2.6. (#5213) 2025-04-10 09:14:38 -07:00
Ke Bao
1078396f47 Update deps for mllama4 (#5215) 2025-04-10 09:12:44 -07:00
Teng Ma
7e4f72dd8c [PD] Add get_contiguous_buf_infos interface for MLATokenToKVPool (#5204) 2025-04-10 20:05:34 +08:00
Teng Ma
4c31ae9f6d [PD] Support KV transfer with mooncake (#4880)
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
Co-authored-by: Shangming Cai <caishangming@linux.alibaba.com>
Co-authored-by: Xuchun Shang <xuchun.shang@linux.alibaba.com>
Co-authored-by: shangmingc <csmthu@gmail.com>
2025-04-10 14:23:23 +08:00
Xiaoyu Zhang
f730362ee2 reduce moe_align_block_size_kernel small batch mode overhead (#5086) 2025-04-09 17:59:35 -07:00
fzyzcjy
e3c4bd3153 Fix DeepSeek error when using DeepEP mode (#5190) 2025-04-09 17:43:22 -07:00
Stefan He
5db37c8626 [metrics] Add in queue metrics (#4444) 2025-04-09 17:19:27 -07:00
Yineng Zhang
4cb53ecd0c fix: log warning when disable cuda graph (#5209) 2025-04-09 14:16:13 -07:00
Zhaoyang Hao
456b008bd8 Add H20 dtype fp8_w8a8 fused MoE kernel tuning configs for DeepSeek V3/R1 (#5196) 2025-04-09 11:54:36 -07:00
Yi Zhang
ebf495f013 sgl-kernel use cutlass latest version for fp8 blockwise gemm (#5207) 2025-04-09 11:47:04 -07:00
saienduri
7f875f1293 update grok test (#5171) 2025-04-09 11:09:47 -07:00
Mick
fbebcb7aa4 model: support mllama4 (#5144) 2025-04-09 09:28:44 -07:00
Xiaoyu Zhang
87eddedfa2 [ci] fix ci test fused_moe op (#5102) 2025-04-09 08:52:46 -07:00
HandH1998
4065248214 Support Llama4 fp8 inference (#5194)
Co-authored-by: laixinn <xielx@shanghaitech.edu.cn>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
Co-authored-by: zhyncs <me@zhyncs.com>
2025-04-09 20:14:34 +08:00
fzyzcjy
86a876d883 Optimize topk operation in llama4 (#5128) 2025-04-09 02:50:22 -07:00
kk
92823069c4 Fix ci test "test_eval_fp8_accuracy" failed (#5185)
Co-authored-by: wunhuang <wunhuang@amd.com>
2025-04-09 02:44:05 -07:00
yinfan98
d2e507df3c [Misc] clean up vllm in sgl-kernel test (#5189) 2025-04-09 01:22:13 -07:00
fzyzcjy
61970b08d8 Let bench_one_batch support enable_dp_attention (#4058) 2025-04-08 23:44:25 -07:00
Cheng Wan
76c48a0913 [DeepEP] fix: import buffer error (#5179) 2025-04-08 22:12:14 -07:00
Yineng Zhang
90caf06c00 fix: use DeepEPDispatcher on CUDA (#5180) 2025-04-08 21:56:53 -07:00
Yineng Zhang
6669d12707 feat: add DeepGEMM build warning (#5176)
Co-authored-by: grimoire <streetyao@live.com>
2025-04-08 21:16:23 -07:00
Kay Yan
f2b70afde0 docs: remove the use of Downward API for LWS_WORKER_INDEX (#5110)
Signed-off-by: Kay Yan <kay.yan@daocloud.io>
2025-04-08 20:46:11 -07:00
Jinyan Chen
bc3f6db2dd [Fix] DeepEP Compatibility with Low Latency (#5068)
Co-authored-by: ch-wan <cwan39@gatech.edu>
2025-04-08 20:31:31 -07:00
Chang Su
aac531c53b [Bugfix] Fix index out of bounds in local attention with large sequences (#5173) 2025-04-08 18:43:13 -07:00
fzyzcjy
39efad4fbc Tiny disable model that does not work (#5175) 2025-04-08 18:42:37 -07:00
fzyzcjy
466899e69c Fix multimodal hashing error (#5174) 2025-04-08 18:42:26 -07:00