Commit Graph

4586 Commits

Author SHA1 Message Date
DangKai
e970892ffa fix import qwenvl error in RL engine (#12874) 2025-11-16 12:01:16 -08:00
b8zhong
d5fa58c4dd fix nightly docker build (#13386) 2025-11-16 11:21:09 -08:00
Liangsheng Yin
e019f233f9 Remove unused code / testcases in lang (#13335) 2025-11-16 22:48:35 +08:00
Xinyuan Tong
b1c688fba2 refactor: cleanup vision attention related codes (#13228)
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: alisonshao <54658187+alisonshao@users.noreply.github.com>
Co-authored-by: Mick <mickjagger19@icloud.com>
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
Co-authored-by: Kangyan-Zhou <zky314343421@gmail.com>
2025-11-16 22:44:40 +08:00
Zilin Zhu
8e3663d4e8 Add default enable_memory_saver to HybridLinearKVPool (#13371) 2025-11-16 22:35:13 +08:00
Zilin Zhu
9edb0e0da3 Add SGLANG_ENABLE_REQ_POOL_LEAK_STRICT_CHECK to bypass mem leak check (#13339) 2025-11-16 22:02:22 +08:00
Xiaoyu Zhang
95f43669b5 [2 / 2] apply sgl-kernel weak_ref_tensor (#12978)
Co-authored-by: Yuan Luo <yuan.luo@hotmail.com>
2025-11-16 21:24:35 +08:00
Xiaoyu Zhang
50691d7b49 [opt kimi k2 2/n] apply kimi k2 thinking moe_fused_gate (#13332) 2025-11-16 21:20:08 +08:00
Yuhao Yang
6afe396399 diffusion: support fa4 in fa backend for blackwell (#13263)
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-11-16 21:02:45 +08:00
Adarsh Shirawalmath
d724670873 [diffusion] refactor and added tests for Flux, T2V, TI2V, I2V(#13344) 2025-11-16 19:58:04 +08:00
Zilin Zhu
9509c4ccd3 [RL] enable offloading hybrid linear attn model (#13336) 2025-11-16 17:12:52 +08:00
Qiaolin Yu
78a4b446c6 Fix dpsk-r1-fp4 tp8 by reverting two commits (#13162 and #13341) (#13348)
Co-authored-by: Kangyan-Zhou <zky314343421@gmail.com>
2025-11-15 21:31:36 -08:00
DarkSharpness
f969664172 [Performance] Move the contiguous to torch compile region (#13199) 2025-11-15 20:49:52 -08:00
b8zhong
f35f7f1245 [Piecewise CUDA Graph] Support W4A8 (#13179) 2025-11-16 11:53:50 +08:00
ash-sigh
12c789ebd1 [Ascend]support xgrammar backend for ascend npu (#12310)
Co-authored-by: ronnie_zheng <zl19940307@163.com>
2025-11-15 19:39:24 -08:00
kyleliang-nv
597d416070 [feature] Add layerwise NVTX support (#11870) 2025-11-15 19:20:56 -08:00
sglang-bot
1ca205f6da chore: bump sgl-kernel version to 0.3.17.post1 (#13358) 2025-11-15 19:11:41 -08:00
b8zhong
24a25ffa20 [Piecewise CUDA Graph] Support ModelOpt FP4 (#13101) 2025-11-16 11:03:19 +08:00
Lianmin Zheng
db7299aa30 [Fix] Register custom ops only if they exist (#13321) 2025-11-15 17:48:17 -08:00
Vincent Zhong
0d116b9a0b Clean up deprecated tile_tokens_dim for next flashinfer (#13341)
Signed-off-by: Vincent Zhong <207368749+vincentzed@users.noreply.github.com>
2025-11-16 00:50:15 +08:00
FlyPanda
51f9b9628b [optimize] Provide Usrbio compilation and installation commands (#12329) 2025-11-15 23:06:24 +08:00
xlzheng
37e8724ef3 perf: optimize TypeBasedDispatcher using dict for O(1) lookup (#12001) 2025-11-15 22:02:42 +08:00
Zhihao Lyu
10592e9c08 [Ascend][Feat] Add Ascend sampling backend (#12692)
Co-authored-by: ronnie_zheng <zl19940307@163.com>
2025-11-15 21:59:58 +08:00
Even Zhou
2aec8b6e1b [Feature] Spec-Overlap supporting DP-ATTN; PD-Disaggregation; npugraph mode (#12443) 2025-11-15 21:51:07 +08:00
Adarsh Shirawalmath
3f400f25df [Diffusion] add health endpoints to diffusion server (#13329) 2025-11-15 19:12:37 +08:00
Yongfei Xu
d91b16eb16 Opt tp: tp attn support tp reduce scattered input (#10568) 2025-11-15 02:08:12 -08:00
Charles Chen
b051d76dab Fix: add missing get_embed_and_head in MiniMax M2 for Eagle3 (#13297) 2025-11-15 01:36:03 -08:00
fzyzcjy
d971f22898 Super tiny expose transform_scale_ue8m0 API for RL frameworks (#13323) 2025-11-15 17:31:04 +08:00
Zilin Zhu
bc083521a8 [RL] support update_weights_from_tensor for mtp (#7415) 2025-11-15 16:57:01 +08:00
fzyzcjy
33f08a98b0 Tiny refactor condition to requant scale ue8m0 (#13286) 2025-11-15 16:36:00 +08:00
fzyzcjy
8e6083bfcf Support inverse transform ue8m0 scale (#13285) 2025-11-15 16:34:32 +08:00
fzyzcjy
2fbc78a083 Support fast gemm when in batch invariant DeepGEMM fallback (#13259) 2025-11-15 16:34:15 +08:00
Zilin Zhu
f0b5ccf5f5 [RL] Allow bypassing /health check (#13320) 2025-11-15 16:33:54 +08:00
Zilin Zhu
56fc483073 [RL] support only do cpu backup on draft model (#13318) 2025-11-15 16:31:57 +08:00
Minglei Zhu
8a4373405e re-submit 12911 but relax the requirement for deepgemm (#13226) 2025-11-15 15:37:12 +08:00
Yuan Luo
f0021c0dc8 Add feature flag for mm inputs processing optimization (#13278)
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
2025-11-15 14:43:44 +08:00
Douglas Yang
7ee3e36412 Fix: test_vlm_offline_throughput output throughput (#13279) 2025-11-15 14:31:48 +08:00
Ho-Ren (Jack) Chuang
6d5e16fb1c feat: Add FP4 (E2M1) KV Cache Support for MHA (#12612)
Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
Co-authored-by: Yichen Wang <yichen.wang@bytedance.com>
2025-11-14 22:31:35 -08:00
Zilin Zhu
34851471b2 Add more statistics for spec decoding (#13317) 2025-11-15 14:20:28 +08:00
Adarsh Shirawalmath
c2083116e3 [Diffusion] switch to local calculate_dimensions (#13294) 2025-11-15 13:55:13 +08:00
Yiming
9b3fc186bc Fix syntax errors in cpp_radix_tree (#13315) 2025-11-15 11:35:43 +08:00
Adarsh Shirawalmath
af373636da CI: add server performance test for SGLang diffusion (#13091)
Co-authored-by: Mick <mickjagger19@icloud.com>
2025-11-15 10:11:19 +08:00
Minglei Zhu
a5be6ef98e [Deterministic] Support Qwen3-Next model deterministic inference (#13100) 2025-11-14 18:01:18 -08:00
Yuwei An
eae59b337e Piecewise Cuda Graph Support for gpt-oss model (#13045)
Signed-off-by: Oasis-Git <ayw.sirius19@gmail.com>
2025-11-15 09:28:00 +08:00
alisonshao
dcc47a56c9 Implement nightly test workflow naming conventions (#13170) 2025-11-14 16:14:53 -08:00
Kaixi Hou
5ae0ac4244 [NVIDIA] Fix use case of SGLANG_ENABLE_FLASHINFER_GEMM (#13274)
Co-authored-by: Baizhou Zhang <sobereddiezhang@gmail.com>
2025-11-14 12:51:11 -08:00
Lianmin Zheng
22f641ab4f [minor] remove debug code in python/sglang/srt/compilation/weak_ref_tensor_jit.py (#13235)
Co-authored-by: Stefan He <hebiaobuaa@gmail.com>
2025-11-14 12:22:24 -08:00
Stefan He
0997c78d2c Support FP8 Per Token Quant Piecewise (#13272) 2025-11-14 11:40:12 -08:00
Praneth Paruchuri
665f43bdd8 model: support teleflm (#10573) 2025-11-15 03:14:49 +08:00
Praneth Paruchuri
a53f2d6c12 Support orion (#10665) 2025-11-15 03:08:32 +08:00