-
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
[AMD] Register 4 JIT kernel unit tests for AMD CI
run-ci
#27837
opened Jun 10, 2026 by
michaelzhang-ai
Collaborator
Loading…
2 tasks done
[Metrics] Fix
fwd_occupancy reading NaN on every decode log line
run-ci
#27836
opened Jun 10, 2026 by
hnyls2002
Collaborator
Loading…
[bugfix][AMD] Disable aiter allreduce+RMSNorm fusion under DP attention / EP
#27835
opened Jun 10, 2026 by
okorzh-amd
Loading…
5 tasks
[AMD] Enable BCG on ROCm + route aiter prefill via MHA during PCG/BCG…
#27833
opened Jun 10, 2026 by
karverma-amd
Loading…
DeepSeek-V4 support disaggregation-decode-enable-radix-cache and support MTP
#27831
opened Jun 10, 2026 by
zhangxiaolei123456
Contributor
Loading…
5 tasks
Optimize FLUX.1 tensor parallel sharding
diffusion
SGLang Diffusion
run-ci
#27826
opened Jun 10, 2026 by
mickqian
Collaborator
Loading…
Support Diffusion Gemma
documentation
Improvements or additions to documentation
Multi-modal
multi-modal language model
#27823
opened Jun 10, 2026 by
kpham-sgl
Collaborator
Loading…
2 of 5 tasks
[AMD] ci: add label-gated extra-a tier (kv_canary + mock_model self-tests)
amd
npu
run-ci
run-ci-extra
#27822
opened Jun 10, 2026 by
michaelzhang-ai
Collaborator
Loading…
3 of 4 tasks
[AMD] Gate quark mxfp4 kv_b_proj post-processing on per-layer quant method
deepseek
#27821
opened Jun 10, 2026 by
ColinZ22
Contributor
Loading…
3 tasks done
Revert "[AMD] Fix DeepSeek V4 Pro c128 state tensor dtype mismatch (#27529)" — fp32 compress perf regression
jit-kernel
#27820
opened Jun 10, 2026 by
DarkSharpness
Collaborator
Loading…
[NPU] Add proxy for accessing https://raw.githubusercontent.com
#27819
opened Jun 10, 2026 by
e-martirosian
Contributor
Loading…
5 tasks
[AMD] ci: register 9 attention-backend unit tests to run on AMD CI
run-ci
#27817
opened Jun 10, 2026 by
michaelzhang-ai
Collaborator
Loading…
2 of 3 tasks
Update test_aiter_allgather_amd.py
amd
bypass-fastfail
run-ci
#27815
opened Jun 10, 2026 by
kangwangamd
Contributor
Loading…
5 tasks
[DO NOT MERGE] Revert routed_scaling_factor fusion — Nemotron CI debug for release
quant
LLM Quantization
#27812
opened Jun 10, 2026 by
b8zhong
Collaborator
Loading…
[AMD] Restore AMD piecewise CUDA graph support dropped by #23906
run-ci
#27811
opened Jun 10, 2026 by
fxmarty-amd
Contributor
Loading…
1 task done
bugfix for npu mtp graph runner
npu
run-ci
#27808
opened Jun 10, 2026 by
Hexq0210
Contributor
Loading…
5 tasks
[DeepSeek-V4] feat: W4A8 MXFP4 MoE backend for DeepSeek-V4 on SM90 (FlashInfer)
#27806
opened Jun 10, 2026 by
yuan-luo
Collaborator
Loading…
4 tasks
[HiCache] Support draft KV pool for UnifiedRadixCache(DeepSeekV4)
#27805
opened Jun 10, 2026 by
kevincheng2
Loading…
[AMD] Fix HIP fallback modulation math
amd
diffusion
SGLang Diffusion
#27804
opened Jun 10, 2026 by
nonam3e
Loading…
5 tasks done
bugfix revise interface get cpu copy for npu mem pool to align with gpu
npu
#27802
opened Jun 10, 2026 by
McZyWu
Contributor
Loading…
5 tasks
fix(tokenizer_manager): log full text on finish in incremental streaming (#27775)
#27801
opened Jun 10, 2026 by
Anai-Guo
Loading…
Integrate InstantTensor into SGLang
dependencies
Pull requests that update a dependency file
documentation
Improvements or additions to documentation
#27800
opened Jun 10, 2026 by
arlo-scitix
Loading…
5 tasks done
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.