Skip to content

Pull requests: sgl-project/sglang

New pull request New
184 Open 2,913 Closed
184 Open 2,913 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Automatically retry when HuggingFace Hub raises 429 HTTP error
#4845 opened Mar 28, 2025 by fzyzcjy Loading…
6 tasks
[WIP] Add FA3 draft backend for speculative decoding high priority
#4841 opened Mar 28, 2025 by zcnrex • Draft
6 tasks
@zhyncs
test: reduce mem_fraction_static for gemma3 vision test
#4840 opened Mar 28, 2025 by vhain Loading…
6 tasks
@merrymercy @zhyncs
16
Fix DeepSeek V3 cannot run on 4x8xH100 high priority
#4836 opened Mar 28, 2025 by fzyzcjy Loading…
6 tasks
@zhyncs
Clean up import vllm in quantization/__init__.py high priority
#4834 opened Mar 28, 2025 by merrymercy Loading…
@fzyzcjy @zhyncs
6
Support Page Size > 1 for FA3 high priority
#4832 opened Mar 27, 2025 by hebiao064 Loading…
1 of 7 tasks
@qingquansong @zhyncs @Fridge003
[WIP] Support FA3 backend for MLA high priority
#4831 opened Mar 27, 2025 by Fridge003 • Draft
1 of 8 tasks
@zhyncs
4
[WIP] Add FA3 into CMAKELIST high priority
#4824 opened Mar 27, 2025 by hebiao064 • Draft
6 tasks
@yinfan98
add __half22bfloat162 __bfloat1622half2 for awq_kernel with cu121 SDK
#4820 opened Mar 27, 2025 by yiakwy-xpu-ml-framework-team Loading…
1 of 6 tasks
@sleepcoo
5
[sgl-kernel] per token group quant support COLUMN MAJOR
#4817 opened Mar 27, 2025 by BBuf Loading…
2 of 8 tasks
3
DeepEP full DP, change activation communication to FP8.
#4800 opened Mar 26, 2025 by sleepcoo • Draft
@ch-wan
10
Modified docs
#4796 opened Mar 26, 2025 by simveit • Draft
1
Tiny fix doc error
#4795 opened Mar 26, 2025 by fzyzcjy Loading…
6 tasks
enable vlm use fast image procesor
#4794 opened Mar 26, 2025 by yizhang2077 Loading…
6 tasks
@mickqian
6
Support non-attention path operators in Triton
#4792 opened Mar 26, 2025 by JackChuang Loading…
3 of 6 tasks
@Alcanderian
11
Improving Total Token Throughput by 1%: Reducing CPU Overhead in Zero-Overhead Scheduling
#4790 opened Mar 26, 2025 by WANG-GH Loading…
1
Fix AttributeError in scheduler's release_memory_occupation method
#4789 opened Mar 26, 2025 by GeLee-Q Loading…
6 tasks
Fix wrong variable name when stopping memory profile
#4772 opened Mar 25, 2025 by Fr4nk1inCs Loading…
6 tasks
2
[Benchmark] Tilelang DeepGEMM implementation with dynamic block_N & num_stages
#4769 opened Mar 25, 2025 by zcnrex • Draft
6 tasks
bump sgl-kernel 0.0.5.post4
#4768 opened Mar 25, 2025 by zhyncs Loading…
6 tasks
1
[Feature] Support DeepEP Low Latency high priority
#4767 opened Mar 25, 2025 by liz-badada Loading…
1 of 6 tasks
@ch-wan
2
support sharded for remote model load-format
#4763 opened Mar 25, 2025 by AllenXu93 Loading…
1
Add integration test for Flash Attention 3
#4760 opened Mar 25, 2025 by yubofredwang Loading…
2 of 6 tasks
5
fix typo: disagg_prefill_infight_queue -> disagg_prefill_inflight_queue
#4757 opened Mar 25, 2025 by GaoYusong Loading…
1 of 6 tasks
refactor: multimodal data
#4754 opened Mar 25, 2025 by mickqian Loading…
6 tasks
2
ProTip! Mix and match filters to narrow down what you’re looking for.