Skip to content

Pull requests: sgl-project/sglang

New pull request New
1,243 Open 10,759 Closed
1,243 Open 10,759 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

Support pre-generating and using expected checksums
#16730 opened Jan 8, 2026 by fzyzcjy Loading…
5 tasks
7
Support bitwise weight checksum verifier
#16729 opened Jan 8, 2026 by fzyzcjy Loading…
5 tasks
8
Adapt cann 8.5: use sfa and lightning indexer op from cann npu
#16728 opened Jan 8, 2026 by randgun Loading…
5 tasks
1
[Fix] Apply --dist-timeout to GroupCoordinator process groups
#16726 opened Jan 8, 2026 by luoyuyan Loading…
5 tasks
1
support smem in per_token_quant_fp8 kernel quant LLM Quantization sgl-kernel
#16725 opened Jan 8, 2026 by zhangxin81 Loading…
5 tasks
5
[Rework] Add SwapAB Optimization for triton fused_moe_kernel on SM90. run-ci
#16723 opened Jan 8, 2026 by Insideyyy Loading…
1 of 5 tasks
4
[AMD] Change AITER package name amd
#16721 opened Jan 8, 2026 by yctseng0211 • Draft
5 tasks
1
[diffusion] model: wan tp+usp optimize diffusion SGLang Diffusion run-ci
#16720 opened Jan 8, 2026 by triple-mu Loading…
5 tasks
13
[Diffusion] model: fix zimage tp diffusion SGLang Diffusion run-ci
#16719 opened Jan 8, 2026 by CPFLAME Loading…
5
Skip causal_conv1d test with padded batches due to Triton kernel bug
#16715 opened Jan 8, 2026 by alisonshao Loading…
6
[AMD CI] migrate test_deepseek_r1_mxfp4_8gpu to new ci registry amd deepseek
#16713 opened Jan 8, 2026 by bingxche • Draft
5 tasks
3
Add force-include-usage Support for stream
#16711 opened Jan 8, 2026 by syd520zy Loading…
2 of 5 tasks
2
Fix AMD CI suite names in pr-test-amd.yml amd run-ci
#16710 opened Jan 8, 2026 by alisonshao Loading…
3
[smg][ci] migrate chat completions tests to new infrastructure and build wheel once and share via artifact model-gateway run-ci
#16709 opened Jan 8, 2026 by slin1237 Loading…
5 tasks
1
fix spec qwen3 pd error run-ci
#16708 opened Jan 8, 2026 by whybeyoung Loading…
1
5
[Rocm][Feat] Accelerate VisionAttention by precompute H2D part in every vis… Multi-modal multi-modal language model
#16705 opened Jan 8, 2026 by ZLkanyo009 Loading…
1
[diffusion] feat: support decode with large num frames diffusion SGLang Diffusion
#16704 opened Jan 8, 2026 by zcnrex Loading…
5 tasks
3
[PCG]Add print tensor op for debug piecewise-cuda-graph
#16702 opened Jan 8, 2026 by Chen-0210 Loading…
1 task done
1
qwen3_vl encoder support graph Multi-modal multi-modal language model npu quant LLM Quantization
#16700 opened Jan 8, 2026 by chenxu214 Loading…
5 tasks
1
feat(ascend): add runtime support for GPTQ-quantized MoE models + AutoRound-quantized dense and MoE models
#16699 opened Jan 8, 2026 by GuoYechang Loading…
2
fix(function_call): group batch decode by options instead of fallback
#16698 opened Jan 8, 2026 by ruokee Loading…
1 of 5 tasks
1
1
Add some more metrics documentation Improvements or additions to documentation
#16694 opened Jan 8, 2026 by vincentzed • Draft
5 tasks
1
fix: auto-append <｜Assistant｜> when last message is not user/dev
#16689 opened Jan 8, 2026 by joeyhacker Loading…
1
[Quantization][NPU] support gptq moe layer on npu
#16688 opened Jan 8, 2026 by 22dimensions • Draft
5 tasks
1
fix none lora eviction
#16687 opened Jan 8, 2026 by ConnorLi96 Loading…
5 tasks
1
ProTip! Filter pull requests by the default branch with base:main.