-
Notifications
You must be signed in to change notification settings - Fork 6.8k
Pull requests: sgl-project/sglang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[AMD][DI][CI] 2/N Add DSV4 DP8/EP8 and MTP MI355X 1P1D nightly recipes
run-ci
#29784
opened Jun 30, 2026 by
Lzy17
Contributor
Loading…
3 tasks done
Potential fix for NVFP4 accuracy
deepseek
#29783
opened Jun 30, 2026 by
b8zhong
Collaborator
Loading…
[AMD] Register 4 CPU-bound unit/mem_cache + utils tests for AMD 1-GPU PR CI
run-ci
#29782
opened Jun 30, 2026 by
michaelzhang-ai
Collaborator
Loading…
3 tasks
[AMD] Enable EAGLE for GLM5.2-MXFP4
deepseek
#29781
opened Jun 30, 2026 by
Raiden-Makoto
Contributor
Loading…
2 of 5 tasks
[LPLB] Replace hand-written cholesky with cuSolverDx::posv for IPM
jit-kernel
#29780
opened Jun 30, 2026 by
dorispnvidia
•
Draft
5 tasks
Share one logits output buffer across prefill/decode/draft cuda-graph runners
run-ci
run-ci-extra
#29779
opened Jun 30, 2026 by
cctry
Collaborator
Loading…
[diffusion] Support SP for Krea-2
diffusion
SGLang Diffusion
documentation
Improvements or additions to documentation
#29777
opened Jun 30, 2026 by
AgainstEntropy
Collaborator
Loading…
4 of 5 tasks
[DeepSeek V4] Enable FlashMLA sparse prefill by default
#29775
opened Jun 30, 2026 by
YAMY1234
Collaborator
Loading…
3 tasks done
[diffusion] Shard QwenImage DiT across TP ranks
diffusion
SGLang Diffusion
#29774
opened Jun 30, 2026 by
decajoin
Contributor
Loading…
2 of 5 tasks
Add class-level defaults on ModelRunner for attributes set in load_model()
#29773
opened Jun 30, 2026 by
davislx
Loading…
2 of 5 tasks
[multimodal_gen] Fix USPAttention replicated-prefix head sharding for GQA
diffusion
SGLang Diffusion
run-ci
#29772
opened Jun 30, 2026 by
AgainstEntropy
Collaborator
Loading…
4 of 5 tasks
[MoE] Route ungrouped softmax/sigmoid topk through the unified Triton router
jit-kernel
run-ci
run-ci-extra
#29771
opened Jun 30, 2026 by
BBuf
Collaborator
Loading…
4 of 5 tasks
chore: prune low-value cleanup code
amd
dependencies
Pull requests that update a dependency file
diffusion
SGLang Diffusion
documentation
Improvements or additions to documentation
model-gateway
mthreads
Multi-modal
multi-modal language model
npu
quant
LLM Quantization
sgl-kernel
[CI] Relax diffusion CI thresholds
diffusion
SGLang Diffusion
run-ci
run-ci-extra
#29767
opened Jun 30, 2026 by
BBuf
Collaborator
Loading…
[WIP] [MoE] mxfp on a5 initial support
deepseek
documentation
Improvements or additions to documentation
npu
quant
LLM Quantization
#29762
opened Jun 30, 2026 by
OrangeRedeng
Contributor
•
Draft
5 tasks
[Bugfix] compressed-tensors WNA16 MoE: don't assume a "Linear" config group
run-ci
#29761
opened Jun 30, 2026 by
joerowell
Contributor
Loading…
5 tasks
Remove transformers 5.12.1 dead-code workarounds
run-ci
#29758
opened Jun 30, 2026 by
JustinTong0323
Collaborator
Loading…
Fix fake-transfer prefill: chunked KV leak + radix cache exclusion
#29757
opened Jun 30, 2026 by
imReese
Loading…
[AMD] Fix MiniMax M3 state transfer in Mori PD
#29756
opened Jun 30, 2026 by
YukioZzz
Loading…
5 tasks
[Diffusion] cache cross-attn K/V across denoise steps for Helios
diffusion
SGLang Diffusion
#29755
opened Jun 30, 2026 by
LLThomas
Contributor
Loading…
3 of 5 tasks
[HiCache] fix: MooncakeStore(standalone) check fails for DeepSeek-V4 logical anchor pool
#29754
opened Jun 30, 2026 by
mitu626
Loading…
2 of 5 tasks
Previous Next
ProTip!
no:milestone will show everything without a milestone.