-
Notifications
You must be signed in to change notification settings - Fork 6.5k
Pull requests: sgl-project/sglang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[NPU] Add NPU fallback for fused Triton gating kernels
run-ci
#28293
opened Jun 15, 2026 by
iridiumine
Contributor
Loading…
2 of 5 tasks
[Fix] model init / XPU / transformers-v5 / bench-image fixes
#28292
opened Jun 15, 2026 by
vshekhawat-hlab
Contributor
Loading…
2 of 5 tasks
[AMD][MXFP4] Reland "Online MXFP4 quantization 2/N - FP8 to MXFP4 requantization on AMD GPUs"
documentation
Improvements or additions to documentation
quant
LLM Quantization
#28291
opened Jun 15, 2026 by
fxmarty-amd
Contributor
Loading…
[AMD] Test DeepSeek V4 FlashMLA backend variants nightly
amd
deepseek
#28290
opened Jun 15, 2026 by
bingxche
Collaborator
Loading…
[Fix] Handle list-typed matched values in trim_matched_stop
#28289
opened Jun 15, 2026 by
playaswd
Loading…
Fix SGLANG_DEFAULT_THINKING ignored by dsv4/dsv32 reasoning parser
#28288
opened Jun 15, 2026 by
WeiLai5432
Contributor
Loading…
[HiCache] Optimize HiCache hash generation with bulk token byte conversion
run-ci
#28287
opened Jun 15, 2026 by
huangtingwei9988
Collaborator
Loading…
5 tasks
[Test] Add unit tests for srt/managers/scheduler_recv_skipper.py and srt/managers/scheduler_input_blocker.py
npu
#28285
opened Jun 15, 2026 by
evanderfff123-boop
Contributor
Loading…
3 of 5 tasks
Fix inaccuracies and add NPU constraints in ascend_npu_profiling.mdx.
documentation
Improvements or additions to documentation
npu
#28283
opened Jun 15, 2026 by
qinsir5522
Loading…
5 tasks
Fix priority-preemption budget dropping prompt tokens for fresh requests
#28282
opened Jun 15, 2026 by
fzyzcjy
Collaborator
Loading…
[AMD] Point AITER scout at amd/aiter-ci
amd
#28281
opened Jun 15, 2026 by
bingxche
Collaborator
Loading…
Allow overriding tokenizer path in benchmark harness
run-ci
#28280
opened Jun 15, 2026 by
merrymercy
Contributor
Loading…
[NPU] fix ascend_docs
documentation
Improvements or additions to documentation
npu
run-ci
#28279
opened Jun 15, 2026 by
Hide-on-bushsh
Contributor
Loading…
5 tasks
[Fix] baichuan: place alibi_slopes on XPU device instead of CUDA
#28278
opened Jun 15, 2026 by
vshekhawat-hlab
Contributor
Loading…
2 of 5 tasks
[HiCache] Add NIXL tenant startup cleanup
documentation
Improvements or additions to documentation
hicache
Hierarchical Caching for SGLang
#28276
opened Jun 15, 2026 by
chivalryq
Contributor
Loading…
5 tasks done
[GLM5][MoE] perf: Avoid copy materialization for trtllm MoE
deepseek
run-ci
#28274
opened Jun 15, 2026 by
mattteochen
Contributor
Loading…
5 tasks
shard DeepGEMM warmup across local GPUs
documentation
Improvements or additions to documentation
#28271
opened Jun 15, 2026 by
shiyu7
Contributor
Loading…
5 tasks done
[Fix] Unblock event loop during Qwen-VL multimodal processing (fixes #28247)
#28270
opened Jun 15, 2026 by
abinggo
Contributor
Loading…
2 tasks done
[Unified Tree]Support host-only HiCache tree
run-ci
#28269
opened Jun 15, 2026 by
huangtingwei9988
Collaborator
Loading…
5 tasks
dsv4 test ci
deepseek
jit-kernel
npu
run-ci
#28268
opened Jun 15, 2026 by
Talantan1102
Loading…
5 tasks
[NPU] Add causal conv1d
npu
#28267
opened Jun 15, 2026 by
zhaozx-cn
Contributor
Loading…
5 tasks done
[Diffusion] enable cache-dit for ERNIE-Image model
diffusion
SGLang Diffusion
#28266
opened Jun 15, 2026 by
LLThomas
Contributor
Loading…
4 of 5 tasks
Fix bug of enable both piecewise_cuda_graph and mamba_extra_buffer
piecewise-cuda-graph
run-ci
#28262
opened Jun 15, 2026 by
ZhouMengLei1999
Loading…
[Bugfix]Fix Qwen3-VL DFlash aux hidden state capture setup
#28261
opened Jun 15, 2026 by
gq112
Contributor
Loading…
4 of 5 tasks
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.