Skip to content

Pull requests: sgl-project/sglang

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
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

[DeepSeek-V32]Update nightly performance benchmark deepseek
#15308 opened Dec 17, 2025 by Fridge003 Loading…
6 tasks
[Deepseek V3.2] Support Overlap Spec + NSA run-ci
#15307 opened Dec 17, 2025 by b8zhong Loading…
4 tasks done
Fix warp illegal instruction in kimi k2 thinking PCG sgl-kernel
#15306 opened Dec 17, 2025 by BBuf Loading…
6 tasks
[2/2] Fused qk_norm_rope for GLM4.6
#15305 opened Dec 17, 2025 by Kevin-XiongC Draft
6 tasks
Fix the accuracy issue when running mxfp4 dsv3 model and enable ep
#15304 opened Dec 17, 2025 by kkHuang-amd Loading…
1 of 6 tasks
[Fix] A followup fix for TRTLLM BF16 MoE run-ci
#15303 opened Dec 17, 2025 by samuellees Loading…
6 tasks
fix: adding retry loop to huggingface weight download
#15300 opened Dec 17, 2025 by dougyster Loading…
1 task
[AMD] Add TP=8 models to nightly test amd
#15296 opened Dec 17, 2025 by michael-amd Draft
6 tasks
fix mindspore import warning documentation Improvements or additions to documentation run-ci
#15287 opened Dec 16, 2025 by b8zhong Loading…
Support per token Int8/Int4 KV Cache diffusion SGLang Diffusion documentation Improvements or additions to documentation quant LLM Quantization reinforcement-learning run-ci
#15285 opened Dec 16, 2025 by jindajia Loading…
1 of 6 tasks
Revert "direct register custom op for mm_fp4 (#13699)" blackwell SM100/SM120 quant LLM Quantization run-ci
#15284 opened Dec 16, 2025 by b8zhong Loading…
Fix FP4 KV cache with EAGLE speculative decoding blackwell SM100/SM120
#15279 opened Dec 16, 2025 by xwuShirley Loading…
feat: DeepSeek-V3.2 Streaming tool call output deepseek diffusion SGLang Diffusion documentation Improvements or additions to documentation run-ci
#15278 opened Dec 16, 2025 by JustinTong0323 Loading…
6 tasks
Reserve more memory for DeepSeekOCR model to reduce flakiness Multi-modal multi-modal language model run-ci
#15277 opened Dec 16, 2025 by Kangyan-Zhou Loading…
6 tasks
Add NFS bandwidth check and AMD CI utilities amd diffusion SGLang Diffusion documentation Improvements or additions to documentation run-ci
#15275 opened Dec 16, 2025 by sunxxuns Loading…
3 tasks
diffusion: Add ComfyUI plugin support for SGLang Diffusion diffusion SGLang Diffusion documentation Improvements or additions to documentation
#15271 opened Dec 16, 2025 by niehen6174 Loading…
6 tasks
eagle3 support dp, and llama support dp
#15268 opened Dec 16, 2025 by chenxu140 Loading…
6 tasks
[diffusion] refactor: deprecate WorkloadType diffusion SGLang Diffusion
#15267 opened Dec 16, 2025 by mickqian Loading…
6 tasks
[Schedule] refactor SchedulerEnhancer to SchedulerPrefillGroup
#15265 opened Dec 16, 2025 by liupeng374 Loading…
6 tasks
MiniMax-M2: pipeline-parallel safe forward/weight-loading + comm
#15264 opened Dec 16, 2025 by jtreal Loading…
6 tasks
[diffusion]Support url image input diffusion SGLang Diffusion
#15262 opened Dec 16, 2025 by IPostYellow Loading…
6 tasks
[AMD] add unit-test-backend-8-gpu-amd back amd
#15253 opened Dec 16, 2025 by yctseng0211 Loading…
6 tasks done
ProTip! Add no:assignee to see everything that’s not assigned.