-
Notifications
You must be signed in to change notification settings - Fork 4.6k
Pull requests: sgl-project/sglang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[diffusion] support torch compile for diffusers backend
diffusion
SGLang Diffusion
#19673
opened Mar 2, 2026 by
DefTruth
Loading…
3 of 5 tasks
support fused_moe_triton and moe_sum_all_reduce kernel fusion[reduce …
#19672
opened Mar 2, 2026 by
xieminghe1
Loading…
5 tasks
[Qwen3.5] Support Qwen3.5 Pipeline Parallelism
performance
run-ci
#19670
opened Mar 2, 2026 by
yuan-luo
Loading…
5 tasks
[HiCache] refactor: hicache normalization flow and compatibility checks
run-ci
#19669
opened Mar 2, 2026 by
alphabetc1
Loading…
5 tasks
[CI] Make Lora tests more robust by changing default dtype to bfloat16
lora
run-ci
#19667
opened Mar 2, 2026 by
Fridge003
Loading…
5 tasks
[CPU] improve numa memory binding
run-ci
sgl-kernel
#19666
opened Mar 2, 2026 by
blzheng
Loading…
5 tasks
[AMD] CI - enable FlyDSL MOE a4w4 support
amd
#19665
opened Mar 2, 2026 by
yctseng0211
•
Draft
5 tasks
[Debug] Add torch._assert_async to Eagle spec decoding gather paths
#19664
opened Mar 2, 2026 by
hnyls2002
Loading…
3 tasks
[Bugfix] DeepSeekV32 tool calls don't support streaming output for arguments when tool_choice="auto"
deepseek
#19662
opened Mar 2, 2026 by
Huixxi
Loading…
2 of 5 tasks
[Mamba]: Refactor additional_ratio calculation when init mamba pool
run-ci
#19660
opened Mar 2, 2026 by
hzh0425
Loading…
5 tasks
[diffusion] postprocess: Fix CI in frame interplation number of frames
diffusion
SGLang Diffusion
run-ci
#19659
opened Mar 2, 2026 by
yyy1000
Loading…
5 tasks
[Bugfix] For cp: Fixed hang problem in prefix cache and kvcache support fp8 in-seq-split mode
#19656
opened Mar 2, 2026 by
Baidu-AIAK
Loading…
Update sgl-attn to include SWA decode optimizations
run-ci
sgl-kernel
#19655
opened Mar 2, 2026 by
zminglei
Loading…
[diffusion] Initial XPU support for sglang diffusion
dependencies
Pull requests that update a dependency file
diffusion
SGLang Diffusion
intel
run-ci
xpu
intel gpu with device `torch.xpu`
#19653
opened Mar 2, 2026 by
xiangyuT
Loading…
1 of 5 tasks
[Feature] NVFP4 Marlin fallback for non-Blackwell GPUs (SM75+)
blackwell
SM100/SM120
quant
LLM Quantization
#19652
opened Mar 2, 2026 by
Godmook
Loading…
5 tasks done
[diffusion] postprocess support for upscaling
diffusion
SGLang Diffusion
documentation
Improvements or additions to documentation
#19648
opened Mar 2, 2026 by
jiangyukunok
•
Draft
5 tasks
Re-land sync patch with median KL fix
high priority
run-ci
#19646
opened Mar 2, 2026 by
alisonshao
Loading…
1 of 2 tasks
[feat] support prompt token ids return without recompute prompt logprobs
#19645
opened Mar 2, 2026 by
guapisolo
Loading…
5 tasks
[BUG] Support tuple hidden_states from fused MXFP4/FP8 quantization
run-ci
#19643
opened Mar 2, 2026 by
zyzshishui
Loading…
5 tasks
[TestFix] change LoRA tests to use NVIDIA adapter instead of Nutanix
documentation
Improvements or additions to documentation
lora
run-ci
#19642
opened Mar 2, 2026 by
glenliu21
Loading…
3 tasks done
Remove sync points in logits_processor: async H2D transfers
#19640
opened Mar 2, 2026 by
YazhiGao
Loading…
3 tasks
[Bugfix] Fix the bug blocking the startup of Llama-3.2-11b
#19638
opened Mar 2, 2026 by
xdtbynd
Loading…
3 of 5 tasks
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-02-02.