Commit Graph

5494 Commits

Author SHA1 Message Date
Xinyuan Tong
aab35bccb4 fix: draft model IMA by overide max_positional_embeddings (#10787)
Co-authored-by: Qiaolin Yu <qy254@cornell.edu>
2025-09-23 12:56:16 -07:00
Yineng Zhang
42aedb02af [Auto Sync] Update protocol.py (20250923) (#10820)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: cicirori <32845984+cicirori@users.noreply.github.com>
2025-09-23 12:49:56 -07:00
Yiakwy
984730b732 add tunning files for QWEN-3-NEXT (#10794) 2025-09-23 12:46:30 -07:00
Shangming Cai
23632d350c Fix latest main ci (#10799)
Signed-off-by: Shangming Cai <csmthu@gmail.com>
2025-09-23 12:46:13 -07:00
Chang Su
08b8c0c3cd [router] fix axum default body limit (#10818) 2025-09-23 12:44:17 -07:00
Lzhang-hub
d42975c641 Remove duplicate code in qwen2 model (#10540) 2025-09-24 02:40:51 +08:00
ZhengHSI
adc24a3a0c fix ceval (#10504)
Co-authored-by: lukec <118525388+sleepcoo@users.noreply.github.com>
2025-09-24 02:35:25 +08:00
Chang Su
7ff93e613f router(grpc): Implement route for chat_cmpl endpoint (#10761) 2025-09-23 11:26:33 -07:00
Simo Lin
b24b2e7ed7 [router] use dashmap for radix tree instead of hash for multi model (#10814) 2025-09-23 11:25:53 -07:00
sogalin
7135db5d31 [ROCm] Update aiter to v0.1.5.post3 (#10812) 2025-09-23 10:54:12 -07:00
Jimmy
4b5ef3002c [fix][pd-disag]no need set next batch sampling info done in prefill (#10259) 2025-09-24 01:24:36 +08:00
Zheng Li
4f564b9e83 model: support qwen3-vl series (#10323)
Co-authored-by: ocss884 <ocss.lin@gmail.com>
Co-authored-by: cao1zhg <653506626@qq.com>
Co-authored-by: yhyang201 <yhyang201@gmail.com>
Co-authored-by: yhyang201 <47235274+yhyang201@users.noreply.github.com>
Co-authored-by: 瑀澈 <yuche.lz@alibaba-inc.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
Co-authored-by: Yineng Zhang <me@zhyncs.com>
2025-09-23 10:15:52 -07:00
Simo Lin
98c3b04ff2 [router] responses api POST and GET with local storage (#10581)
Co-authored-by: key4ng <rukeyang@gmail.com>
2025-09-23 09:12:02 -07:00
Simo Lin
ddab4fc7c7 [router] fix cache aware routing strategy and lock contention (#10773) 2025-09-23 08:53:49 -07:00
Shangming Cai
d21c35224d Fix hicache mooncake backend CI (#10792)
Signed-off-by: Shangming Cai <csmthu@gmail.com>
2025-09-23 02:04:44 -07:00
Liangsheng Yin
4a762041d7 move environ into sglang.srt to avoid break SRT auto sync. (#10791) 2025-09-23 02:04:20 -07:00
Lianmin Zheng
ea338676b5 Clean up server args (#10770) 2025-09-23 00:22:32 -07:00
ishandhanani
b06db198ba followup: clean up dockerfiles and release yamls (#10783) 2025-09-23 00:19:46 -07:00
Yineng Zhang
8c1ef0f914 chore: upgrade sgl-kernel 0.3.12 (#10782) 2025-09-23 00:18:54 -07:00
Liangsheng Yin
f5a2faf2b8 Introduce FutureMap (#10715) 2025-09-22 23:27:30 -07:00
ishandhanani
1c82d9db28 feat: unify dockerfiles (#10705)
Co-authored-by: Yineng Zhang <me@zhyncs.com>
2025-09-22 23:23:48 -07:00
Lifu Huang
9241f4fd20 Move cached kernel to srt.utils (#10776) 2025-09-22 23:00:36 -07:00
fzyzcjy
063c3791fe Fix trtllm_mla slow concat kernel in MTP (#10777) 2025-09-22 22:47:49 -07:00
Liangsheng Yin
632b7d8cc9 Use simulate acc len from sglang.environ (#10771) 2025-09-23 12:59:50 +08:00
Chang Su
16adf3dcab [router] fix logger type mismatch (#10774)
Co-authored-by: Simo Lin <linsimo.mark@gmail.com>
2025-09-22 21:02:28 -07:00
Simo Lin
c3a1d7759f [router] remove pd router draining channel (#10767) 2025-09-22 20:49:33 -07:00
Simo Lin
89971c4c3c [router] refactor router and worker management 4/n (#10756)
Co-authored-by: Chang Su <chang.s.su@oracle.com>
2025-09-22 18:35:10 -07:00
Lianmin Zheng
113f8f65a2 [Auto Sync] Update configurer.py (20250923) (#10765)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-09-22 17:34:34 -07:00
ronnie_zheng
e22f3a5ec9 [Ascend]optimize Qwen3 on Ascend (#10574)
Co-authored-by: c30031083 <chenxu140@huawei.com>
2025-09-22 17:18:36 -07:00
ronnie_zheng
095093ee5a [Ascend] optimize Qwen-vl on Ascend (#10556)
Co-authored-by: wangqihui01 <wangqh10@163.com>
2025-09-22 17:18:16 -07:00
Even Zhou
d27a6f7092 [Feature] Add MLAProcess for DeepSeek MLA on NPU (#10130) 2025-09-22 17:17:48 -07:00
Yineng Zhang
0753ef831e [Auto Sync] Update flashattention_backend.py (20250922) (#10762)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Gordon Gustafson <ggustafson@together.ai>
2025-09-22 16:41:42 -07:00
ishandhanani
662393f27d fix: kv events with tp > 1 (#10541) 2025-09-22 15:55:44 -07:00
pranavm-nvidia
b1bb8e7490 Enables TRT-LLM backend to be used for target_verify (#10281)
Co-authored-by: Pranav Marathe <pranavm@ipp1-3309.ipp1a1.colossus.nvidia.com>
Co-authored-by: fzyzcjy <ch271828n@outlook.com>
2025-09-22 15:54:00 -07:00
Lianmin Zheng
38c00ed7a1 Fix multimodal registry and code sync scripts (#10759)
Co-authored-by: cctry <shiyang@x.ai>
2025-09-22 15:36:01 -07:00
pansicheng
d4041a5eeb refactor zero copy (#10300)
Co-authored-by: 晟海 <huangtingwei.htw@antgroup.com>
Co-authored-by: huangtingwei <141888744+huangtingwei9988@users.noreply.github.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
2025-09-22 15:17:31 -07:00
Vedant Jhaveri
2f555c4cee [Generative Score API] Added test_scores_api.py to github CICD to run per commit (#10755)
Co-authored-by: Vedant Jhaveri <vjhaveri@linkedin.com>
Co-authored-by: Sundara Raman Ramachandran <sundar24295@gmail.com>
2025-09-22 14:41:57 -07:00
Yineng Zhang
e53df7c009 chore: bump sgl-kernel v0.3.12 (#10732) 2025-09-22 14:39:25 -07:00
Jue WANG
9c53dad809 Fix MTP MoE weight loading with NVFP4 target model. (#10758) 2025-09-22 14:21:57 -07:00
Simo Lin
7ca1bea63d [router] update ci so only execute benchmarks when labels are added (#10757) 2025-09-22 13:23:07 -07:00
Simo Lin
97c3823931 [router] refactor router and worker management 3/n (#10727) 2025-09-22 12:17:50 -07:00
Chang Su
60dbbd086a bugfix: Fix get_worker_urls_for_model in http/router.rs (#10754) 2025-09-22 14:10:31 -04:00
Baizhou Zhang
aa1c5cf5bd Add warnings and remove dependency for deterministic inference (#10724)
Co-authored-by: Yineng Zhang <me@zhyncs.com>
2025-09-22 10:56:02 -07:00
Adarsh Shirawalmath
592caab66a [Docs, minor] Fix LLM doc matrix (#10753) 2025-09-23 01:29:55 +08:00
Lifu Huang
2101d93b4f Fix CI TestChunkedSGMV (#10737) 2025-09-22 16:09:58 +08:00
Shangming Cai
70e4b21853 Fix flaky logprobs test (#10728)
Signed-off-by: Shangming Cai <csmthu@gmail.com>
2025-09-22 00:46:26 -07:00
feng397
944f1ea0ec fix capture_bs when speculative decoding enabled (#10730) 2025-09-22 00:43:12 -07:00
Yingchun Lai
9d7e82a0ab EPLB: prefer to use physical experts in the same node (#9849) 2025-09-22 00:34:30 -07:00
Yang Yu
f05805515b Convert FLASHINFER_WORKSPACE_SIZE to integer (#10731) 2025-09-22 00:26:42 -07:00
Lifu Huang
635ccda673 [4/4] Introduce CachedKernel to reduce CSGMV kernel launch overheads by 60% (#10709) 2025-09-21 22:26:42 -07:00