minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 10 Packages Projects Releases Wiki Activity
14,886 Commits 2 Branches 0 Tags
e85ef548772e3fc416779eadfd1509f4d31e5db3
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Baizhou Zhang 5eb1b6a7ba Remove retired DSA env paths (#29912) 2026-07-05 22:58:02 -07:00
Baizhou Zhang e552f6ed75 [Fix] Fix DSA indexer fusion for NeoX RoPE (#30111) 2026-07-04 03:20:55 -07:00
Baizhou Zhang 1f0f353d92 [CI] Add GLM52 NVFP4 MTP B200 tests (#30021) 2026-07-03 13:43:38 -07:00
Baizhou Zhang e3258d3b54 [Fix] Turn off dsa indexer fusion by default (#30018) 2026-07-03 01:08:06 -07:00
Baizhou Zhang c312cdd3a7 Upgrading tvm-ffi/sgl-deep-gemm/tilelang (#29554) 2026-07-01 12:32:18 -07:00
Baizhou Zhang 677a11bfa9 [Doc] Tiny update dsv4 doc (#29827) 2026-07-01 01:47:35 -07:00
Baizhou Zhang cc238270b1 Update GLM tests to 5.2 and delete redundant tests (#29686) 2026-06-29 23:04:26 -07:00
Baizhou Zhang 12f76d115c Update GLM-5.2 B300 and GB300 NVFP4 cookbook settings (#29466) 2026-06-26 16:15:33 -07:00
Baizhou Zhang 73741f7074 Bypass legacy GLM DSA layer types validation (#29454) 2026-06-26 15:24:16 -07:00
Baizhou Zhang 3344b73c80 Add DeepSeek V4 Pro GB300 nightly and expand Kimi K25 nightly test (#28103) 2026-06-25 12:13:52 -07:00
Baizhou Zhang e3026ef016 [3/N][CP] Implement zigzag CP strategy (#28421) 2026-06-18 15:10:30 -07:00
Baizhou ZhangandShijin 4b817f5d7f Upgrade fa3 hash (#28394) 
Co-authored-by: Shijin <dovis.zhang02@gmail.com>
 2026-06-17 13:32:11 -07:00
Baizhou Zhang 27291118b9 Upgrade sgl-deep-gemm to 0.1.3 (#28402) 2026-06-16 23:06:36 -07:00
Baizhou Zhang 77f327cb6e [2/n] [CP] Add context parallel strategy abstractions (#27313) 2026-06-16 00:20:04 -07:00
Baizhou Zhang 4c0457f440 [misc] Update codeowner (#28342) 2026-06-15 14:42:01 -07:00
Baizhou Zhang a14d1a5656 Add DeepSeek V4 MTP acceptance length checks (#28098) 2026-06-12 18:53:20 -07:00
Baizhou Zhang d26b90cf70 Add weireweire to CI permissions (#27957) 2026-06-11 14:10:45 -07:00
Baizhou Zhang 125ef88892 Disable async assert in Nemotron nightly tests (#27838) 2026-06-10 16:28:14 -07:00
Baizhou Zhang 3c1b0fb226 [1/n] [CP] Simplify prefill context parallel server args (#27312) 2026-06-10 14:11:38 -07:00
Baizhou Zhang 047e5df3b1 Revert "Share BCG output buffers across capture sizes" (#27758) 2026-06-09 20:51:40 -07:00
Baizhou Zhang 9a48bf75f5 Add GB300 base C CI suite (#27427) 2026-06-06 02:27:30 -07:00
Baizhou Zhang e69cc07a60 Remove DeepSeek V4 release Docker workflow (#27404) 2026-06-05 16:49:05 -07:00
Baizhou Zhang cc67f922cd [misc] Update Codeowner for Lora (#27224) 2026-06-03 22:50:15 -07:00
Baizhou Zhang d097cd2212 [GLM-5] Apply trtllm MHA kernel for GLM-5 on Blackwell (#21332) 2026-06-03 17:16:54 -07:00
Baizhou Zhang 68caf49154 Update sgl-deep-gemm to 0.1.2 (#26993) 2026-06-02 13:51:54 -07:00
Baizhou Zhang 2fdae94e46 docs: update RTX PRO 6000 deployment snippet (#26968) 2026-06-01 14:34:27 -07:00
Baizhou Zhang 05bef58297 [Misc] Tiny change default building branch of sgl-deep-gemm (#26699) 2026-05-29 01:18:36 -07:00
Baizhou Zhang 69362cbc2c [Doc] Update benchmark instruction for dsv4 (#26668) 2026-05-28 23:37:35 -07:00
Baizhou Zhang 435c4ffb30 [CI] Clean DeepSeek V4 tests and installation scripts (#26609) 2026-05-28 15:22:59 -07:00
Baizhou Zhang cd65be98df [CP] Add back qwen 30b test (#26604) 2026-05-28 14:34:41 -07:00
Baizhou ZhangandClaude Opus 4.7 68e5b4fdd6 [NemotronH] Fix weight-loading unit test broken by Puzzle support (#26522) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-27 17:44:45 -07:00
Baizhou ZhangandClaude Opus 4.7 d6032c04b6 [docs] Fix V4 Pro balanced recipe (#26451) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-26 23:50:26 -07:00
Baizhou Zhang 0c34fc5ace [Misc] Update CI Permission (#26435) 2026-05-26 19:57:17 -07:00
Baizhou Zhang f0ba651d66 [Doc] Update pip install commands for Cuda12 (#26344) 2026-05-25 22:28:27 -07:00
Baizhou Zhang 29e245e6a7 [misc] Update permission (#26336) 2026-05-25 19:50:06 -07:00
Baizhou Zhangandyhyang201 7b7f1067bd Add non-MTP DSV4 test coverage (#26141) 
Co-authored-by: yhyang201 <yhyang201@gmail.com>
 2026-05-22 19:43:17 -07:00
Baizhou Zhang 0c8049d9ba Update CI permissions and CODEOWNERS (#25826) 2026-05-19 18:19:21 -07:00
Baizhou Zhang cd012ada58 [Fix] Fix extra uninstall of cutlass packages (#25756) 2026-05-19 10:01:38 -07:00
Baizhou Zhangandhnyls2002 b79e4b1e68 [Fix] Try to fix error caused by latest cutedsl packages (#25690) 
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
 2026-05-18 16:51:32 -07:00
Baizhou Zhang 6f892047ec [misc] Throw error when single batch overlap is enabled on Hopper (#25509) 2026-05-18 14:51:32 -07:00
Baizhou Zhang 6dcacb1159 [Doc] Fix several places for dpsk v4 cookbook (#25506) 2026-05-16 21:54:15 -07:00
Baizhou Zhang af26b71ae8 [Misc] Update release branch cut script (#25468) 2026-05-16 00:51:48 -07:00
Baizhou Zhang 1a1d69507d [Doc] Update MegaMoE usage (#25378) 2026-05-15 02:17:50 -07:00
Baizhou Zhangsglang-botb8zhongmmangkad
 		0c19540550 [Fix] Fix gpt oss triton kernels and upgrade flashinfer back to 0.6.11.post1 (#25335) 
Co-authored-by: sglang-bot <sglang-bot@users.noreply.github.com>
Co-authored-by: b8zhong <b8zhong@users.noreply.github.com>
Co-authored-by: mmangkad <mmangkad@users.noreply.github.com>
 2026-05-15 01:04:56 -07:00
Baizhou Zhang f7efff321d [Docker] Fix several dependencies in Dockerfile (#25215) 2026-05-13 20:18:26 -07:00
Baizhou Zhangandpranjalssh b7f856df70 DeepSeek V4 w4a4 MegaMoE (#25052) 
Co-authored-by: pranjalssh <adkz.photos@gmail.com>
 2026-05-13 18:35:32 -07:00
Baizhou Zhang 51a9403104 Update flashinfer to 0.6.11.post1 (#25129) 2026-05-13 00:12:19 -07:00
Baizhou Zhang ea217a2bf0 ci: remove Execute Notebooks workflow (#24460) 2026-05-11 00:26:32 -07:00
Baizhou ZhangXinyuan Tongyueming-yuan
 		ef5e9f8aba [DSV4] Cherry pick missing commits from deepseek_v4 branch and enhance tests (#24793) 
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
Co-authored-by: yueming-yuan <yym022502@gmail.com>
 2026-05-09 04:15:37 -07:00
Baizhou Zhang c4bb3ce273 Fix stuck when enabling MTP on DSA models (#24635) 2026-05-07 17:06:28 -07:00
Baizhou Zhang ecb786c8d7 [Kernel] Deprecate DeepGemm in sgl kernel and apply custom wheel sgl-deep-gemm (#24268) 2026-05-06 18:59:01 -07:00
Baizhou Zhang 7fa48916f5 Expand support matrix for pypi wheel release (#24565) 2026-05-06 17:39:02 -07:00
Baizhou Zhang 9e1336d406 [Misc] Fix breaking weight checker test (#24553) 2026-05-06 15:42:54 -07:00
Baizhou Zhang 7ec18f7e4e [Doc] Fix instruction on Cuda 13 environments (#24516) 2026-05-06 02:37:23 -07:00
Baizhou ZhangandClaude Opus 4.7 e0af36a907 [Fix] Fix pypi release workflow (#24417) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-05 01:21:06 -07:00
Baizhou ZhangandClaude Opus 4.7 f059e0264a Fix sgl-deep-gemm release workflow (#24385) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-04 14:36:59 -07:00
Baizhou ZhangandClaude Opus 4.7 de08c804c5 Add release workflow for sgl-deep-gemm wheels (#24348) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-04 00:58:44 -07:00
Baizhou Zhang 200944b415 Update kernel installation instructions after shifting default cuda to 13 (#24181) 2026-05-02 16:07:20 -07:00
Baizhou ZhangandClaude Opus 4.7 7bc7775260 [CI] Fix stage-b-test-4-gpu-b200 silently skipped, hanging wait-for-stage-b (#24208) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-30 22:29:08 -07:00
Baizhou ZhangandClaude Opus 4.7 1742bfb610 [CI] release-whl-kernel: skip musa wheels in update_kernel_whl_index (#24191) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-30 17:17:50 -07:00
Baizhou Zhang 5f88c8593a [Misc] Redirect default sglang nightly wheel to cuda 130 (#24183) 2026-04-30 16:32:50 -07:00
Baizhou Zhang b3ead32d3c [minor] Remove incorrect note after supporting w4a16 moe for DeepSeek V4 (#24035) 2026-04-29 13:26:13 -07:00
Baizhou Zhang 4e885baa9b docs(cookbook): add H200 (FP4) deployment option for DeepSeek-V4 (#23980) 2026-04-28 19:38:53 -07:00
Baizhou Zhang c1d1412333 [Flashinfer] Integrate flashinfer router gemm for sm103 (#23285) 2026-04-27 20:37:56 -07:00
Baizhou Zhang 3177fa7951 Enable DeepGemm warmup in DeepSeek-V4 cookbook (#23883) 2026-04-27 18:41:12 -07:00
Baizhou ZhangandClaude Opus 4.7 977830e91e ci(deepseek-v4): add b300/grace-blackwell dev-branch build options (#23778) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-04-26 16:44:16 -07:00
Baizhou Zhang 69485a176c Small udpate gb300 recipe for deepseek v4 (#23690) 2026-04-24 21:35:16 -07:00
Baizhou ZhangandClaude Opus 4.6 c304d0d64d [Refactor] Deduplicate NSA utils.py into cp_utils.py for context parallel (#22914) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-19 21:35:35 -07:00
Baizhou ZhangKangyan ZhouClaude Opus 4.7Alison ShaoMick
 		6ecd6f84db [CI] Add per-job uv venv isolation and upgrade CI version to Cuda 13 (#23119) 
Co-authored-by: Kangyan Zhou <zky314343421@gmail.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: Alison Shao <a.shao@wustl.edu>
Co-authored-by: Mick <mickjagger19@icloud.com>
 2026-04-19 05:32:36 -07:00
Baizhou Zhang 113d654152 [Fix] Fix accuracy bug in Flashmla sparse MLA kernel (#22723) 2026-04-15 13:40:04 -07:00
Baizhou ZhangandClaude Opus 4.6 8fe9bbffb6 [CI] Reinstall flashinfer-jit-cache on CUDA version mismatch (#22741) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-13 23:04:23 -07:00
Baizhou Zhang b441317aa4 Revert "Upgrade CI default CUDA version from 12.9 to 13.0" (#22727) 2026-04-13 14:39:24 -07:00
Baizhou Zhang 2e70e4f4f6 [CI] Little renaming of gb200 CI workflow (#22608) 2026-04-11 17:52:42 -07:00
Baizhou Zhang d14d368191 [Kernel] Set sgl_per_token_group_quant_8bit_v2 as default choice (#22467) 2026-04-11 01:59:57 -07:00
Baizhou ZhangandClaude Opus 4.6 3c46ff2ac5 fix: restore CPU flash_attn test to use sgl_kernel directly (#22573) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-10 21:39:20 -07:00
Baizhou Zhang 60acdc31f2 [Fix] Fix several bugs on DSA models (#22430) 2026-04-09 12:46:23 -07:00
Baizhou Zhang 606aa11ea8 [DSA] Enable all reduce fusion for DSA models (#22390) 2026-04-09 12:42:44 -07:00
Baizhou Zhang 5e1a9834f9 Update ci permission (#22387) 2026-04-08 15:42:06 -07:00
Baizhou ZhangandClaude Opus 4.6 4e5b8cb041 Fix get_version_tag.py to handle dot-separated post versions (#22385) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-08 15:18:22 -07:00
Baizhou Zhang 213af1d4f7 Add CI tests for GLM-5 (#22285) 2026-04-08 01:05:36 -07:00
Baizhou Zhang 20ee59bcfc [Misc] Remove unused cu13 docker release workflow (#22167) 2026-04-05 16:01:02 -07:00
Baizhou Zhang c5fa364b80 [Hotfix] Fix router gemm on sm103 (#22134) 2026-04-05 09:33:14 -07:00
Baizhou Zhang 106baedbfb [Doc] Update GLM-5 instructions in sglang documentation (#21716) 2026-04-05 03:13:07 -07:00
Baizhou Zhang 088203454b [Fix] Fix nightly tests (#22140) 2026-04-05 02:26:42 -07:00
Baizhou Zhang 723ed6c3c4 [CI]Temporary ban auto benchmark tool test (#22138) 2026-04-04 23:18:19 -07:00
Baizhou Zhang bf984ae65d Revert "[Bugfix] Temporarily skip TRTLLM attention on (G)B300 (SM103) to avoid high-concurrency hang" (#22098) 2026-04-04 02:17:19 -07:00
Baizhou Zhang ac1e437f6a Revert "[Feature] JIT activation and update skills (by codex)" (#22078) 2026-04-03 15:04:15 -07:00
Baizhou Zhang 97adf8a290 [misc] Add hint for kernel release trigger (#22036) 2026-04-03 03:31:44 -07:00
Baizhou ZhangandClaude Opus 4.6 98ac40192b [Workflow] Fix kernel release build failures for aarch64 and wheel renaming (#22018) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-03 03:23:03 -07:00
Baizhou Zhang 75de479680 [Misc] Update CI permission (#22014) 2026-04-02 23:37:05 -07:00
Baizhou ZhangandClaude Opus 4.6 5c082c307a [Workflow] Fix kernel release jobs skipped on push events (#22011) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-02 23:03:13 -07:00
Baizhou Zhang 0a709cfe02 [Workflow] Avoid triggering nightly tests in kernel bump workflow (#22010) 2026-04-02 22:40:33 -07:00
Baizhou Zhang efa7b2d5d3 Revert "[MUSA][9/N] Add FA3 attention backend support through MATE (MUSA AI Tensor Engine)" (#22002) 2026-04-02 20:42:13 -07:00
Baizhou ZhangandClaude Opus 4.6 29d8e959d7 [CI] Remove stale Ascend suite entries from test/srt/run_suite.py (#21978) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-04-02 16:47:19 -07:00
Baizhou Zhang c7d03a6215 Revert "Rollback flashmla to older version [1/2]" (#21922) 2026-04-02 00:27:02 -07:00
Baizhou Zhang fbc1f92453 [DSA] Set trtllm kernels as nsa default for Blackwell (#21914) 2026-04-02 00:22:27 -07:00
Baizhou Zhang 5e12c4e08e [DSA] Support trtllm sparse mla kernel for prefill batches (#21783) 2026-04-01 13:55:05 -07:00
Baizhou ZhangandClaude Opus 4.6 f60f2ccc10 [Fix] Fall back to triton MOE for GPT-OSS on Blackwell with driver >= 595 (#21780) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-31 15:52:10 -07:00
Baizhou Zhang d52757fe97 [CI]Remove msgm-en and mmlu tests which cause timeout (#21733) 2026-03-31 01:10:05 -07:00
Baizhou ZhangandClaude Opus 4.6 62a63eeff7 [Fix] Fix weight_loader property assignment for qwen3-next FP8 models (#21662) 
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-30 01:35:59 -07:00