Commit Graph

6380 Commits

Author SHA1 Message Date
Chang Su
ecb9fa14e6 [docker] clean up main dockerfile for router and dev configurations (#12364)
Co-authored-by: ybyang <ybyang7@iflytek.com>
2025-10-30 07:46:09 -07:00
Simo Lin
700daa3463 [router] harmony responses api streaming support (#12395) 2025-10-30 07:30:48 -07:00
Fan Yin
39cee0fed2 [sgl-kernel] upd deepgemm hash to rebased commit (#11960) 2025-10-30 07:19:00 -07:00
Xiaoyu Zhang
04e5b6faa7 Revert "Triton fused_moe_kernel support ep moe tuning" (#12377) 2025-10-30 07:12:06 -07:00
Even Zhou
ce6b17c0f9 [Feature] Support DeepSeek MTP on NPU (#11897)
Co-authored-by: liupeng374 <liupeng374@huawei.com>
2025-10-30 21:56:56 +08:00
Even Zhou
cafebef154 [NPU] bugfix for Qwen3-Next and performance update (#11969) 2025-10-30 21:52:16 +08:00
Liangsheng Yin
73dfd2dfb1 [Test] Enhance radix cache test for spec cases (#12394) 2025-10-30 18:46:28 +08:00
fzyzcjy
df5192cffa Enable fast silu-and-mul-and-quant fused kernel (#11806) 2025-10-30 18:15:39 +08:00
JensenFire
78c43d8847 [Feature] Initial eagle3 support for Deepseek-like models (#12319) 2025-10-30 17:26:09 +08:00
Kangyan-Zhou
7e28c67d19 Fix DeepSeek chat templates to handle tool call arguments type checking (#11700) (#12123) 2025-10-30 16:39:25 +08:00
Tanjiro
3edba9bcc0 [fix] added image token as prefix for deepseek-ocr (#12358) 2025-10-30 16:37:03 +08:00
Xuchun Shang
e5ec976402 [Bug fix][PP] fix deadlock with tie_word_embeddings (#12362)
Signed-off-by: Xuchun Shang <xuchun.shang@gmail.com>
2025-10-30 01:03:28 -07:00
Baizhou Zhang
621dfb8886 Import flash_mla from sgl-kernel (#12135) 2025-10-29 23:54:21 -07:00
fzyzcjy
fb52d35f63 Super tiny fix AMD ci (#12378) 2025-10-29 23:25:18 -07:00
Kangyan-Zhou
2b71531a70 Allow benchmarking tool to handle empty response (#12174)
Co-authored-by: Claude <noreply@anthropic.com>
2025-10-29 20:26:23 -07:00
fzyzcjy
25c5049870 Super tiny add tag for benchmark scripts (#12340) 2025-10-30 11:19:14 +08:00
Makcum888e
8e2ac2e628 [NPU] fix pp_size>1 (#12195) 2025-10-30 11:18:36 +08:00
Yuan Luo
17a57fd862 [Perf] Optimize multimodal mm_inputs process in scheduler (#11910)
Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>
2025-10-30 10:48:10 +08:00
Teng Ma
32438eba45 [Ckpt Engine] feat: new sglang entrypoint support for update (#12216) 2025-10-30 10:39:27 +08:00
FlyPanda
fed02a49c2 [bugfix] fix deepseekvl2 and deepseek_ocr model type conflict (#12050)
Co-authored-by: herta <herta@pplabs.org>
2025-10-29 18:54:32 -07:00
Simo Lin
03b3e89aea [router] Harmony Pipeline: Chat Completion & Responses API with MCP Support (#12153)
Co-authored-by: Chang Su <chang.s.su@oracle.com>
2025-10-29 18:52:56 -07:00
Trevor Morris
9ff9fa7f95 Fuse wk and weight_proj in Indexer for DeepSeekV3.2-FP4 (#12094) 2025-10-29 18:42:56 -07:00
b8zhong
7ed8ba05cb [CI] Add Llama 3.1 8B FP4 to B200 CI (#12182) 2025-10-29 14:56:41 -07:00
Stefan He
df08f346ce Fix Duplicate Classmethod in spec_info.py (#12354) 2025-10-29 13:57:10 -07:00
Chang Su
db15148c8a [router] update router docker to use maturin and build from local (#12350) 2025-10-29 13:00:57 -07:00
Simo Lin
5259becd3c [bug] fix router installation to include additional dependency (#12348) 2025-10-29 12:45:18 -07:00
AichenF
ed1044ac1b support cutlass fp4 kernel in sm120 (#11737) 2025-10-29 12:25:16 -07:00
Chang Su
d717e73ea8 [router] refactor mcp to use LRU and fix pooling bug (#12346) 2025-10-29 12:18:54 -07:00
weiliang
a18161875c Fix Flashinfer Backend for SM120 Usage (#12325) 2025-10-29 11:51:55 -07:00
Minglei Zhu
e39628fd07 [2/2] Deepseek deterministic: support deepseek v3 deterministic inference on 8 x H200 (#12095) 2025-10-29 11:49:04 -07:00
b8zhong
bacb3825fe fix: llama 4 + trtllm gen + fp8 kv cache incompatibility (#12347) 2025-10-29 11:31:02 -07:00
Simo Lin
b53d9e11c6 [bug] fix router pypi license file (#12345) 2025-10-29 10:05:00 -07:00
zejunchen-zejun
8a6838212a [Fix] fix type issue of env flag value MODELOPT_MAX_TOKENS_PER_EXPERT (#11709)
Signed-off-by: zejunchen-zejun <zejun.chen@amd.com>
2025-10-29 09:44:05 -07:00
Xiaoyu Zhang
52694b60da Triton fused_moe_kernel support ep moe tuning (#12343) 2025-10-29 23:16:09 +08:00
Chang Su
400bddf24c [router] fix router release workflow and add build test in PR (#12315) 2025-10-29 08:11:45 -07:00
Feng Su
1e90fe2ed1 [Bug fix] trace: fix import error in mini_lb if sgl-router image does not install sglang (#12338) 2025-10-29 07:45:08 -07:00
Yuhong Guo
caa5d2967c feat: return partial generation results when aborting requests in waiting queue (#11673) 2025-10-29 22:03:00 +08:00
Rain H
750940ae36 Eagle3 DP attention for Qwen3 MoE (#12002) 2025-10-29 20:25:17 +08:00
Liangsheng Yin
42f8ea4030 [Test] Fix session control test (#12336) 2025-10-29 18:28:04 +08:00
Liangsheng Yin
14cbe42fd3 Refactor abortion in event loop (#12312) 2025-10-29 18:25:20 +08:00
Baizhou Zhang
685c06451f [ci] Try fixing broken CIs (#12317) 2025-10-29 01:13:51 -07:00
Liana Koleva
1357397a34 feat: preview filename from tuning_fused_moe_triton.py (#12276)
Co-authored-by: Xiaoyu Zhang <35585791+BBuf@users.noreply.github.com>
2025-10-29 16:12:25 +08:00
hlu1
42e1a72efb [Deepseek V3.2] Enable flashmla_auto with MTP (#12294)
Signed-off-by: Hao Lu <14827759+hlu1@users.noreply.github.com>
2025-10-28 23:51:20 -07:00
b8zhong
83a7c89c3f followup fix for llama 4 trtllm flashinfer backend (#12314) 2025-10-28 22:17:08 -07:00
Yuzhen Zhou
0380ca82ef Add Batch‑Invariant RMSNorm (#12144) 2025-10-28 21:05:57 -07:00
Yingchun Lai
ec92b0cefe EPLB: prefer to use physical experts in the same gpu or node (#10874) 2025-10-28 21:01:11 -07:00
Liana Koleva
e03b6beeb1 doc: improve modelopt error description (#12269) 2025-10-28 20:57:55 -07:00
Yingchun Lai
5e36a0b455 [metrics][EPLB]: Support selected count of physical experts on each GPU (#9825) 2025-10-28 20:56:19 -07:00
Gao016
0297773a2f a tiny fix for support deepseek bf16 weights (#12313)
Co-authored-by: gaochang <gaochang@U-19PX2WQ1-0350.local>
2025-10-28 20:46:44 -07:00
Baizhou Zhang
587deb15a7 [hotfix] Fix pytest not found in CI (#12311) 2025-10-29 11:07:36 +08:00