Skip to content

Pull requests: sgl-project/sglang

New pull request New
3,155 Open 19,928 Closed
3,155 Open 19,928 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[LoRA] DSA indexer targets + MoE-LoRA cuda-graph and RL adapter-reload fixes lora
#29874 opened Jul 1, 2026 by yushengsu-thu Collaborator Loading…
2
docs(GLM-5.1): add --enable-aiter-allreduce-fusion to MXFP4 MI355X command documentation Improvements or additions to documentation
#29873 opened Jul 1, 2026 by jiacao-amd Loading…
1
[Package] Move easydict (LGPL-3.0) to an optional remote-models extra dependencies Pull requests that update a dependency file npu
#29869 opened Jul 1, 2026 by gaurav0107 Loading…
5 tasks done
1
2
[Spec][4/N] Decoupled speculative decoding: ignore_decode_budget for the drafter engine run-ci
#29868 opened Jul 1, 2026 by zhendonghua Contributor Loading…
3
feat(short-conv): shared ShortConvAttnBackend for ZAYA1 CCA + LFM2 short conv bypass-fastfail run-ci
#29867 opened Jul 1, 2026 by ch-wan Collaborator Loading…
9
Fix capture-mode detection during breakable CUDA graph capture run-ci run-ci-extra
#29866 opened Jul 1, 2026 by cctry Collaborator Loading…
3
Support alternative BF16 GEMM deepseek quant LLM Quantization
#29865 opened Jul 1, 2026 by b8zhong Collaborator • Draft
1
[diffusion] Refresh LTX HQ consistency GT diffusion SGLang Diffusion
#29863 opened Jul 1, 2026 by mickqian Collaborator Loading…
1
[diffusion] Add --offload-during-compile to fit max-autotune on tight-memory GPUs diffusion SGLang Diffusion run-ci
#29862 opened Jul 1, 2026 by mickqian Collaborator • Draft
3 tasks
2
Fix SWA eviction tombstoning the last leaf bypass-fastfail run-ci run-ci-extra
#29860 opened Jul 1, 2026 by ispobock Collaborator Loading…
5
[HiCache] Unified Mooncake Registration for Logical Anchors and Draft Pools
#29859 opened Jul 1, 2026 by stmatengss Collaborator Loading…
5 tasks
4
Build SWA window kv buffers for the EAGLE draft-extend cuda-graph path
#29858 opened Jul 1, 2026 by reger-men Loading…
Fix BF16 routing bias dtype for TRT-LLM MoE deepseek
#29856 opened Jul 1, 2026 by milesial Loading…
4
[AMD][DI][CI] 3/N Add Kimi K2.6 FP8 MI355X 1P1D nightly recipes
#29855 opened Jul 1, 2026 by Lzy17 Contributor Loading…
4
docs: complete production metrics reference documentation Improvements or additions to documentation
#29854 opened Jul 1, 2026 by anencore94 Loading…
3
bugfix for npu Grok2 model --detokenizer without all special ids
#29853 opened Jul 1, 2026 by McZyWu Contributor Loading…
5 tasks
2
[multimodal] Refactor CUDA attention backend resolver diffusion SGLang Diffusion
#29852 opened Jul 1, 2026 by mickqian Collaborator • Draft
2
Support MOSS-Transcribe-Diarize model and adapter
#29850 opened Jul 1, 2026 by CloudRipple Contributor Loading…
2 of 5 tasks
3
Experiment: materialize MoE weights before copy amd
#29848 opened Jul 1, 2026 by bingxche Collaborator • Draft
4
[Feature] Add DSA CP shared KV cache documentation Improvements or additions to documentation
#29847 opened Jul 1, 2026 by taoyuanyuan • Draft
5 tasks done
5
[CI] Restore B200 NVFP4 thresholds and sanitize CUDA env dependencies Pull requests that update a dependency file diffusion SGLang Diffusion run-ci run-ci-extra
#29844 opened Jul 1, 2026 by BBuf Collaborator Loading…
3
[trtllm_mha] Fuse cuda-graph metadata rebuild into one triton kernel blackwell SM100/SM120 run-ci
#29843 opened Jul 1, 2026 by pranjalssh Contributor Loading…
@ch-wan @Qiaolin-Yu
1
pad customized_info for mixed output batches bypass-fastfail run-ci run-ci-extra
#29842 opened Jul 1, 2026 by yaof20 Contributor Loading…
1
feat: integrate hpc-ops attention and MoE backends
#29839 opened Jul 1, 2026 by kyang669 Loading…
4 of 5 tasks
9
Fix GLM-DSA raw config restoration
#29837 opened Jul 1, 2026 by san-tian Loading…
3
ProTip! Adding no:label will show everything without a label.