Skip to content

Pull requests: sgl-project/sglang

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

Fix default_max_tokens compute error in responses api when mtp is opened
#18932 opened Feb 17, 2026 by LuYanFCP Loading…
2 of 4 tasks
Fix NSA FP8 KV layout mismatch under MHA one-shot quant LLM Quantization run-ci
#18931 opened Feb 17, 2026 by mmangkad Loading…
5 tasks
[AMD] Unit tests for mtp in GLM-4.7
#18930 opened Feb 17, 2026 by almaslof Loading…
5 tasks
[Fix] Enable Pipeline Parallelism support for Kimi K2.5
#18925 opened Feb 17, 2026 by ieBoytsov Loading…
1 of 5 tasks
[NPU] [Quantization] w4a4 MoE layer support npu
#18924 opened Feb 17, 2026 by OrangeRedeng Draft
2 of 7 tasks
[bugfix?] update outdated unittest document documentation Improvements or additions to documentation
#18919 opened Feb 17, 2026 by SoluMilken Loading…
5 tasks done
[TorchAO] Enable TorchAO LinearMethod and TorchAOConfig
#18916 opened Feb 17, 2026 by ZhiweiYan-96 Loading…
5 tasks
[Test] add unit test for skipping already preempted request
#18912 opened Feb 17, 2026 by glenliu21 Loading…
3 tasks done
feat: add cuda core dump CI warpper run-ci
#18909 opened Feb 16, 2026 by hnyls2002 Loading…
Fix FlashInfer autotune deadlock with --enable-symm-mem
#18908 opened Feb 16, 2026 by alisonshao Loading…
1 task
[Temporarily unblock spec v2 qwen3.5]
#18906 opened Feb 16, 2026 by vincentzed Draft
5 tasks
[jit kernel] Support per_token_group_quant_8bit jit kernel quant LLM Quantization run-ci
#18905 opened Feb 16, 2026 by yuan-luo Loading…
5 tasks
[HiCache] feat: L3 prefetch prometheus metrics documentation Improvements or additions to documentation
#18898 opened Feb 16, 2026 by vladnosiv Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.