kvcache-ai/sglang
1
0
Fork 0
mirror of https://github.com/kvcache-ai/sglang.git synced 2026-07-07 07:47:25 +00:00
Code Issues Packages Projects Releases Wiki Activity
9,885 Commits 157 Branches 0 Tags
cc63c99f112f781605017db018fced686cdc94de
Commit Graph

6437 Commits

Author SHA1 Message Date
Mohammad Miadh Angkad
05ab110e02 [Performance] Force split_k=1 for MXFP4 Triton kernels on Hopper (#16014) 
Signed-off-by: Mohammad Miadh Angkad <MAngkad.BSDSBA2027@aim.edu>
 2026-01-08 23:52:22 +08:00
cheng peng
82a8d77bc0 [Diffusion] model: fix zimage tp (#16719) 
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2026-01-08 22:47:21 +08:00
Xiaoyu Zhang
294ff71d18 [Diffusion] Avoid cpu2gpu sync in flashinfer rope and apply flashinfer rope to wanvideo (#16668) 
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-01-08 22:44:38 +08:00
fzyzcjy
b6e8a0d851 Add file size hints in bitwise model file verifier (#16735) 2026-01-08 22:28:52 +08:00
luoyuyan
fb7609f1dd Fix FP8 MoE NaN with DeepGEMM on Blackwell (#16622) 2026-01-08 22:24:12 +08:00
yuhao
d2ea44f775 VLM: enhance VL embedding model with video input support and revise warm-up strategy (#16635) 
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-01-08 22:12:01 +08:00
fzyzcjy
83abecd0c2 Support pre-generating and using expected checksums (#16730) 2026-01-08 20:26:13 +08:00
fzyzcjy
d54f0a10b4 Support bitwise weight checksum verifier (#16729) 2026-01-08 20:19:25 +08:00
Lianmin Zheng
fb04e7e3c8 [Auto Sync] Update schedule_batch.py, common.py, eagle_info... (20260105) (#16519) 
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: cctry <shiyang@x.ai>
 2026-01-08 02:35:46 -08:00
chenxu214
7dd679cbb9 [NPU][Bugfix] Fix qwen3 error when enable-dp-lm-head (#16115) 2026-01-08 15:15:43 +08:00
陈一涵
48b8dcd42e [jit kernel] support dtype as a cpp template parameter (#16452) 2026-01-08 13:54:33 +08:00
Fan Lin
41b434a7e6 [diffusion] endpoint: add API endpoint to query loaded LoRA adapters information (#16533) 2026-01-08 13:51:26 +08:00
Hubert Lu
4935344fcd [AMD] Fix aiter page-size handling, DeepSeek MLA tuple inputs, and HiCache/FA3 decode-backend override (#16531) 2026-01-07 21:14:32 -08:00
Alison Shao
63cc97f4ef ci: migrate 2-GPU tests to test/registered/ (#16529) 2026-01-07 20:28:16 -08:00
hw-csong
261860e17b [NPU][Bugfix] move free_page logics to cpu (#16608) 2026-01-08 12:00:18 +08:00
Yuhao Yang
e14f5ec8a8 [diffusion] refactor: eliminate redundant parameters in req (#16505) 
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-01-08 11:14:03 +08:00
siyu
6b3f93c4dd vlm: support SGLANG_MM_SKIP_COMPUTE_HASH for bypassing multimodal feature hashing (#16555) 2026-01-08 11:10:00 +08:00
Xiaoyu Zhang
5a5cece561 [Diffusion] clean useless and buggy set_seq_parallel_pg in yunchang (#16669) 2026-01-08 11:08:34 +08:00
Junrong Lin
eec7dbd31e remove redundant max_running_reqs calculation in r3 (#16629) 2026-01-08 09:45:21 +08:00
Changyi Yang
bb798a1c26 [diffusion] fix: reduce default text length for Qwen-Image from 1024 to 512 (#16445) 2026-01-08 09:32:21 +08:00
Baizhou Zhang
38dc5839dd [1/n]deepseek_v2.py Refactor: attention backend handlers and forward method definition (#16306) 2026-01-08 09:22:31 +08:00
Harish
156d97b219 Fix KeyError when logprobs=false in completions endpoint (#16095) 2026-01-07 15:49:02 -08:00
b8zhong
24b30f7757 MoE Refactor: Refactor fp8.py -> flashinfer_trllm.py (#15151) 
Co-authored-by: Brayden Zhong <b8zhong@users.noreply.github.com>
 2026-01-07 15:35:00 -08:00
Alison Shao
0c474273c5 Fix gpt_oss_common import path and migrate core tests (#16426) 2026-01-07 12:58:32 -08:00
Michael
3e73e12458 Revert "Add SwapAB Optimization for triton fused_moe_kernel on SM90." (#16676) 2026-01-07 11:24:47 -08:00
Insideyyy
ee4d2287ab Add SwapAB Optimization for triton fused_moe_kernel on SM90. (#15712) 2026-01-07 23:45:35 +08:00
Yi Zhang
7fc12e0bfa support page size large than 64 for mamba radix cache (#16657) 
Co-authored-by: Hanming Lu <hanming@x.ai>
 2026-01-07 22:52:24 +08:00
Li Jinliang
4d902c8211 [diffusion] bench: upgrade multimodal benchmarks for diverse applications and create a prettier, more intuitive logger. (#16179) 2026-01-07 22:10:23 +08:00
JiLi
fd16c91cb8 Handle Marlin weight restoration and shape recording (QAT INT4 Rollout Part1) (#15238) 
Co-authored-by: Gao016 <yngao016@163.com>
Co-authored-by: yefei12 <xjtu_yefeichen@163.com>
Co-authored-by: yzlnew <yzlnew@gmail.com>
Co-authored-by: Peng Zhang <aniz1905@gmail.com>
 2026-01-07 21:11:24 +08:00
Xiaoyu Zhang
32a6540afc [Diffusion] Fix Ulysses/Ring process group construction under TP to enable correct Wan2.2 tensor parallelism (#16532) 
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-01-07 20:52:36 +08:00
Hudson Xing
98a107d491 Re-enable temp_prefill_info assertion after pairing fix (#16203) 2026-01-07 18:05:17 +08:00
YC Tseng
48381c3b6d [AMD] suppress warning for amd (#16620) 2026-01-07 01:37:40 -08:00
Minglei Zhu
4c85f9d039 Only allocate encoder metadata for encoder-decoder models (#16527) 2026-01-06 22:17:04 -08:00
Baizhou Zhang
7d757d6f17 Clean Some Environment Variables for DeepSeek V32 (#15938) 2026-01-07 14:00:16 +08:00
Chi McIsaac
f066036c8b [diffusion] fix: fix ZImage SP sharding for 5D latents and unpad frames (#16418) 
Signed-off-by: Chi <chixie.mcisaac@gmail.com>
 2026-01-07 13:24:04 +08:00
Yingchun Lai
913b688f21 fix: fill a meaningful tool_index (#16504) 2026-01-06 20:00:09 -08:00
Yuan Luo
53846746bf [VLM] Fix CUDA IPC OOM (#16118) 
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
 2026-01-07 11:30:35 +08:00
MOHENOO
534ac384db [HiCacheStorage & PD] fix prefill bootstrap request host memory leaks (#15439) 
Co-authored-by: yangjia1 <yangjia1@kingsoft.com>
 2026-01-07 11:03:57 +08:00
Mick
badcd02896 [diffusion] chore: automatically enable dit_layerwise_offload for Wan (#16499) 2026-01-07 10:22:08 +08:00
fzyzcjy
d874c8bba4 Tiny support http headers in bench serving (#16606) 2026-01-07 10:15:17 +08:00
fzyzcjy
9a21d89c5b Tiny add metrics for prefill delayer (#16603) 2026-01-07 09:53:52 +08:00
Kangyan-Zhou
18e2ef09d7 Add v1/models endpoint to diffusion model APIs so that they can be discovered by model gateway (#16425) 2026-01-06 14:28:24 -08:00
Alison Shao
0cbd8f3247 ci(test): migrate OpenAI server tests to registered CI system (#16326) 
Co-authored-by: Kangyan-Zhou <zky314343421@gmail.com>
 2026-01-06 11:09:37 -08:00
fzyzcjy
ed307a40bf Tiny add --gsp-ordered (#16575) 2026-01-06 22:43:52 +08:00
Ke Bao
02722b9113 Fix evict swa for overlap scheduler and page size > 1 (#16507) 2026-01-06 22:29:40 +08:00
fzyzcjy
f959250f76 Tiny support sglang_routing_keys_active in engine (#16570) 2026-01-06 22:10:10 +08:00
Junrong Lin
bc2f40bebc [test] Add mamba cache release/resume memory test (#14215) 2026-01-06 15:51:10 +08:00
lw9527
2724b1100b fix double Unicode escape issue in streaming tool_calls parameters (#13518) 
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
 2026-01-05 23:47:26 -08:00
Xiaoyu Zhang
5e5b1183ed [Diffusion] Ring Attention support sage backend (#16496) 2026-01-06 14:45:53 +08:00
jiapingW
1d7ad4afcc [diffusion] fix: fix bench_serving always use prompt fixed prompts (#16201) 
Co-authored-by: jiapingW <root@sgl-training-ray.datacrunch.io>
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-01-06 14:39:07 +08:00