-
Notifications
You must be signed in to change notification settings - Fork 6.4k
Pull requests: sgl-project/sglang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
DeepSeek-V4 shared_expert and ep normal using fused swiglu and quant
deepseek
jit-kernel
quant
LLM Quantization
#27561
opened Jun 8, 2026 by
ckl117
Loading…
5 tasks
[diffusion] Add SANA-WM realtime consistency coverage
diffusion
SGLang Diffusion
jit-kernel
#27560
opened Jun 8, 2026 by
mickqian
Collaborator
Loading…
Add the get_cpu_copy() and get_cpu_copy() functions for deepseekv4 wh…
deepseek
#27559
opened Jun 8, 2026 by
BJWang-ant
Contributor
Loading…
docs: add AI Badgr hosted GPU launch option
documentation
Improvements or additions to documentation
#27558
opened Jun 8, 2026 by
michaelmanly
Loading…
2 of 5 tasks
[HiCache][UnifiedTree] L3 SWA periodic checkpoint
#27557
opened Jun 8, 2026 by
vladnosiv
Contributor
Loading…
Fix off-by-one in vocab boundary NaN guard (use >= vocab_size)
#27556
opened Jun 8, 2026 by
fzyzcjy
Collaborator
Loading…
[AMD] update ROCm AITER commit
amd
#27555
opened Jun 8, 2026 by
bingxche
Collaborator
Loading…
1 task done
fix: preserve divisible FP8 block K configs on CUDA
#27553
opened Jun 8, 2026 by
lmyybh
Loading…
1 of 5 tasks
[Spec] Rename spec-v2 token resolver; fix draft-runner comment
#27552
opened Jun 8, 2026 by
hnyls2002
Collaborator
Loading…
[dLLM] Make FDFO a framework capability for all dLLM algorithms
documentation
Improvements or additions to documentation
#27551
opened Jun 8, 2026 by
Hayden727
Loading…
4 tasks done
[Fix] Avoid applying cuda graph input-buffer registry on non-cuda devices
run-ci
#27549
opened Jun 8, 2026 by
ZailiWang
Contributor
Loading…
DO NOT MERGE - CI sandbox for stateless scheduler b (main merged)
bypass-fastfail
run-ci
run-ci-extra
#27548
opened Jun 8, 2026 by
fzyzcjy
Collaborator
Loading…
[SMG] Fix load imbalance issue of cache_aware when decode is faster than prefill
model-gateway
#27547
opened Jun 8, 2026 by
SYChen123
Contributor
Loading…
5 tasks
fix(pd): do not abort when req.disagg_prefill_dp_rank is used
#27546
opened Jun 8, 2026 by
lawrence-harmonic
Contributor
Loading…
[Fix] mamba: add XPU support for causal_conv1d kernel dispatch
#27544
opened Jun 8, 2026 by
vshekhawat-hlab
Contributor
Loading…
[gemma4] Compute RMSNorm in fp32 to fix deep-layer hidden-state divergence vs HF
#27540
opened Jun 8, 2026 by
yihao-liang
Loading…
[AMD] DO NOT MERGE - test pr27529 dsv4
deepseek
jit-kernel
#27539
opened Jun 8, 2026 by
yctseng0211
Collaborator
•
Draft
5 tasks
[MUSA] bump torchada version to 0.1.59 and workaround PCG limitation.
dependencies
Pull requests that update a dependency file
mthreads
sgl-kernel
#27537
opened Jun 8, 2026 by
yafengio
Contributor
Loading…
5 tasks
fix(sgl-kernel/rocm): honor explicit AMDGPU_TARGET over auto-detection
amd
sgl-kernel
#27535
opened Jun 8, 2026 by
Anai-Guo
Loading…
1 task done
[Intel GPU] Enable fused_experts in fp8.py for quantized models on XPU
intel
run-ci
run-ci-extra
xpu
intel gpu with device `torch.xpu`
#27533
opened Jun 8, 2026 by
polisettyvarma
Contributor
Loading…
5 tasks
Add output of error reasons for router exceptions
model-gateway
#27532
opened Jun 8, 2026 by
BJWang-ant
Contributor
Loading…
[Diffusion] Add SANA-WM with streaming support
diffusion
SGLang Diffusion
jit-kernel
#27531
opened Jun 8, 2026 by
AgainstEntropy
Collaborator
Loading…
3 of 5 tasks
[AMD] Fix DeepSeek V4 Pro c128 state tensor dtype mismatch error and c4_sparse_raw_indices attribute error in cuda graph phase
deepseek
jit-kernel
run-ci
#27529
opened Jun 8, 2026 by
At1a8
Contributor
Loading…
4 of 5 tasks
Previous Next
ProTip!
Exclude everything labeled
bug with -label:bug.