minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity
13,053 Commits 2 Branches 0 Tags
75427c9ca426326f3b65ee97b1d13129f67ca5d2
Commit Graph
8720 Commits
Author SHA1 Message Date
Xiaoyu Zhang 75427c9ca4 Route concat MLA to JIT and remove unused downcast (#25843) 2026-05-23 14:30:43 +08:00
xiaobochen-amdandfanxingran fd3e11973b [AMD][aiter] Fix cuda_graph_kv_indices OOB under page_size>1 (#24587) 
Co-authored-by: fanxingran <fanxingran@amd.com>
 2026-05-22 23:19:46 -07:00
Shangming Cai a241659d18 [PD] Consolidate shared logic into common backend (#25979) 
Signed-off-by: Shangming Cai <csmthu@gmail.com>
 2026-05-23 10:41:20 +08:00
Mick c8cea6d4aa [diffusion] feat: auto-select vae channels_last_3d (#26121) 2026-05-23 10:20:30 +08:00
weireweire 629b6c6a85 correct allreduce fusion and dummy_run alignment in SCATTERED MLP mode (moe_dense_tp_size=1) (#19918) 2026-05-22 19:18:25 -07:00
Cheng WanCheng WanClaude Sonnet 4.6
 		d226f75669 [refactor] unify cuda-graph capture/replay across attention backends (#26134) 
Co-authored-by: Cheng Wan <cheng.wan@radixark.ai>
Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
 2026-05-22 18:51:13 -07:00
Yueming Yuan 208397affc [RL] [Spec v2] Use stop-aware seqlen for returned topk metadata (#26126) 2026-05-22 18:13:16 -07:00
Qiaolin Yu c112f7623a Skip init_mha_chunk_metadata in trtllm_mla when not needed (#26017) 2026-05-22 16:34:16 -07:00
nvjullin cadfa2d025 Support piecewise CUDA graph with NSA (#23351) 2026-05-22 14:39:50 -07:00
maocheng23 2df9e8b4b3 [perf] DeepSeekV3: drop redundant FP32 upcasts in trtllm MoE paths (#25189) 2026-05-22 14:23:57 -07:00
Minglei Zhu b73278e4e0 [Fix]: BCG support for RadixLinearAttention (Qwen3.5 / linear-attn hybrid models) (#25110) 2026-05-22 13:30:35 -07:00
Liangsheng Yin 763174fa6c FutureMap: debug-assert that gather sees a stashed value (#26108) 2026-05-22 13:10:57 -07:00
Kangyan-ZhouandClaude Opus 4.7 085777210c feat(kv-events): expose structured KV-event publisher block on /server_info (#25844) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-23 01:59:45 +08:00
Jialin Ouyangandjialino 06c23d55b5 perf: migrate Req token-id storage to array.array('q') in Scheduler (#25098) 
Co-authored-by: jialino <jialino@fb.com>
 2026-05-22 10:51:07 -07:00
fanghao 5e9bd21979 fix(serving_chat): catch TypeError from tojson on Jinja2 Undefined variables (#20700) 2026-05-22 16:43:26 +00:00
HuangJi 9bbd519f34 [diffusion] fix: fix Wan channels_last_3d VAE decode corruption (#25985) 2026-05-22 23:34:41 +08:00
zhangtao2-1andzhangtao 6baa859a86 Add MiniCPM5 tool call parser for XML-style function calls (#25600) 
Co-authored-by: zhangtao <zhangtao2@modelbest.cn>
 2026-05-22 23:09:40 +08:00
Heyang Huang 8c916a715c [diffusion] feat: support role-based component loading and stage affinity (#25168) 2026-05-22 18:50:23 +08:00
kousakawangandwanghanpei e1dcbca220 [FIX][1/2] fix step3-vl/deepseek-ocr image processor error (#24701) 
Co-authored-by: wanghanpei <wanghanpei@bytedance.com>
 2026-05-22 18:39:47 +08:00
Polisetty V R K Jyothendra Varma 80680dc3fe [Intel GPU] 1/N Fix tilelang import in deepseek v4 rope as optional (#25128) 2026-05-22 18:23:18 +08:00
Liangsheng Yin 10751a4f0c drop FutureIndices wrapper class (#26085) 2026-05-22 02:44:42 -07:00
Yueming Yuan bd6c7e713c [fix] Fallback DeepGEMM activation for unsupported shapes (#26025) 2026-05-22 09:01:40 +00:00
johnnycxmandxinyue.fan cf5f496183 [MUSA][22/N] ci(musa): repack wheels with +musa metadata, refine path filters, sync multimodal tests, and add nightly workflow (#25074) 
Co-authored-by: xinyue.fan <xinyue.fan@mthreads.com>
 2026-05-22 16:24:25 +08:00
Yuhao YangandKe Bao c9153da5dc Fix SWA double-free in disagg decode with MTP speculation (#25805) 
Co-authored-by: Ke Bao <ispobaoke@gmail.com>
 2026-05-22 00:28:54 -07:00
Chi McIsaac a7555fc997 [diffusion] fix: fix MOVA DAC bf16 on ROCm (#25674) 2026-05-22 15:18:49 +08:00
Liangsheng Yin c4b6b5ea1e [core] step 2: drop seq_lens sentinel; SB maintains GPU as seq_lens_cpu mirror (#26020) 2026-05-22 00:12:51 -07:00
Yuhao YangandYaochen Han fa6f4dfb35 improve: combine vit calls for images from different reqs from one batch (#25910) 
Co-authored-by: Yaochen Han <814073252@qq.com>
 2026-05-22 14:57:41 +08:00
fzyzcjy 3f0814974c Fix disaggregation bootstrap server lifetime (#25982) 2026-05-22 14:10:57 +08:00
Khoa PhamandClaude Opus 4.7 f829cafa3e [perf] skip add_special_tokens=False kwarg on chat-template tokenize for slow tokenizers (#25953) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-21 22:45:54 -07:00
Alex O. P. ae7c4226eb [diffusion] model: support FLUX.2-klein-base (#25661) 2026-05-22 11:24:46 +08:00
Liangsheng Yin 8b473aa0bc [core] step 1: route non-spec seq_lens via FutureMap with per-mode bootstrap fixes (#25944) 2026-05-21 20:15:51 -07:00
longguoandYang 4374789abf fix(mm): make multimodal data loading non-blocking to prevent health check stalls (#24751) 
Signed-off-by: abinggo <107740309+abinggo@users.noreply.github.com>
Co-authored-by: Yang <ID+Y-aang@users.noreply.github.com>
 2026-05-22 10:08:03 +08:00
Mick 16b3edc84f [diffusion] feat: enable warmup for sglang serve by default (#25988) 2026-05-22 08:54:48 +08:00
Mick f6d98a17ba [diffusion] optimize: reuse cached dynamic lora weights (#25893) 2026-05-22 08:54:01 +08:00
nohup 7cf193fe1f feat: support HybridLinearKVPool in chunked prefix cache handling (#25753) 2026-05-21 16:17:26 -07:00
Cheng WanandClaude Sonnet 4.6 d765dfd043 refactor(attn): init hisparse_coordinator before attn_backend; replace lazy property with init-time capture (#26012) 
Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
 2026-05-21 16:03:42 -07:00
Cheng WanandClaude Sonnet 4.6 c5251a98a9 feat(model_runner): remove pool/backend refs from ForwardBatch via ForwardContext (#25983) 
Co-authored-by: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
 2026-05-21 14:01:49 -07:00
Liangsheng Yin 44ec2ee18d [core] Unify output_tokens_buf in FutureMap (#25922) 2026-05-21 13:56:44 -07:00
Liangsheng Yin c9a0e55eb5 [Spec] Polish FutureMap after #25879: rename callback, async guard, cleanup (#25962) 2026-05-21 13:56:22 -07:00
Jimmy Shong 1a85586738 [Fix]: Restrict Kimi-K2.5 shared-experts fusion to Quark MXFP4 checkpoints (#25974) 2026-05-21 13:07:45 -07:00
Cheng WanandClaude Opus 4.7 b765faee30 [MoE Refactor] deprecate forward_npu and NpuFuseEPMoE (#25678) 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 2026-05-21 10:25:17 -07:00
Ethan (Yusheng) SuandCursor a24c374f84 [lora] Remove synchronous .any().item() guard in LoRA MoE prefill path (#25531) 
Co-authored-by: Cursor <cursoragent@cursor.com>
 2026-05-21 23:58:57 +08:00
Mick ca9dc17be4 [diffusion] chore: adjust layer wise-offload strategy (#25930) 2026-05-21 23:48:58 +08:00
liuxianglong17 32f996b75a Avoiding the problem of printing a large number of compatibility warn… (#25956) 2026-05-21 22:10:23 +08:00
Kangyan-ZhouandLiangsheng Yin caa9f08294 [CI] Force-reinstall nvidia-cutlass-dsl-libs-cu13 last to avoid wheel-mix TypeError (#25958) 
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
 2026-05-21 22:01:42 +08:00
Liangsheng Yin baeac179f7 [Spec] Route seq_lens through FutureMap; drop verify_done.wait (#25879) 2026-05-21 01:51:40 -07:00
DarkSharpnessandClaude 19f55c0e6d [Refactor] major JIT kernel clean up for dsv4 (#25884) 
Co-authored-by: Claude <noreply@anthropic.com>
 2026-05-21 01:14:31 -07:00
hxie c3f9bc9818 Fix nixl mla key and backup skipping (#24376) 2026-05-21 00:48:28 -07:00
看海的人 b9ae8353d2 [NPU] Support model DeepSeek-OCR and DeepSeek-OCR-2 (#25257) 2026-05-21 15:21:20 +08:00
Liwansi 190488e9a8 [NPU] Support chunk prefill for Qwen3.5/Qwen3.6 models (#25839) 2026-05-21 14:44:25 +08:00