Skip to content

Pull requests: sgl-project/sglang

New pull request New
743 Open 6,683 Closed
743 Open 6,683 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Quick Fix: fix Qwen3-VL launch failure caused by MRotaryEmbedding arg run-ci
#10985 opened Sep 27, 2025 by yhyang201 Loading…
4 tasks
1
[Benchmark] Minor improve for profiling: add output_dir to start_profile in requests
#10984 opened Sep 27, 2025 by Muqi1029 Loading…
1 of 4 tasks
1
fixed pipeline gpt-oss errors run-ci
#10983 opened Sep 27, 2025 by sunxxuns Loading…
1 of 4 tasks
1
[Generative Score API] Multi-Item scoring with custom attention mask. high priority run-ci
#10979 opened Sep 27, 2025 by sundar24295s Loading…
1 of 4 tasks
@hebiao064 @qingquansong @hnyls2002
[router] basic mcp support for openai router response api run-ci
#10978 opened Sep 27, 2025 by key4ng Loading…
4 tasks
8
Fix profiling tools run-ci
#10977 opened Sep 27, 2025 by Qiaolin-Yu Loading…
4 tasks
5
[Ci Monitor] Auto uploaded performance data to sglang_ci_data repo run-ci
#10976 opened Sep 27, 2025 by BBuf Loading…
6
Clean mem settings in server_args and io_struct.py run-ci
#10975 opened Sep 27, 2025 by merrymercy Loading…
2
Add RequestMetricsExporter utility to export request-level metrics
#10973 opened Sep 26, 2025 by scottjlee • Draft
4 tasks
fix: KimiK2Detector Improve tool call ID parsing with regex high priority ready-to-merge The PR is ready to merge after the CI is green. run-ci
#10972 opened Sep 26, 2025 by JustinTong0323 Loading…
4 tasks
5
fix file and object naming scheme in HiCacheNixl to avoid data corruption run-ci
#10969 opened Sep 26, 2025 by ziruiliu Loading…
4 tasks
@xiezhq-hermann @hzh0425
6
[WIP] Add model cache for AMD tests run-ci
#10966 opened Sep 26, 2025 by Kangyan-Zhou Loading…
4 tasks
5
[bugfix]Add empty_context import to two_batch_overlap.py
#10964 opened Sep 26, 2025 by wejoncy Loading…
4 tasks
3
Reworked fast_pos_embed_interpolate() using torch
#10959 opened Sep 26, 2025 by terfendail Loading…
4 tasks
3
refactor global_server_args_dict from dict to dataclass
#10957 opened Sep 26, 2025 by frank285714 Loading…
4 tasks
5
[Test] Fix XPU pytest crashes by initializing mem_fraction_static and setting profiler env
#10954 opened Sep 26, 2025 by singhalshubham03 Loading…
4 tasks
2
Opt MHA chunked prefix: merge prefix and extend kv cache to run mha once run-ci
#10953 opened Sep 26, 2025 by xu-yfei Loading…
1 of 4 tasks
4
Tests/fix attributeerror streaming moe
#10952 opened Sep 26, 2025 by singhalshubham03 Loading…
4 tasks
3
Support encoder_decoder on cpu_graph_runner
#10950 opened Sep 26, 2025 by CaoE • Draft
1 of 3 tasks
Feature/longbench v2 evaluation utils
#10949 opened Sep 26, 2025 by alhridoy Loading…
1
@Fridge003
13
Fix Qwen3-VL-235B-A22B-Thinking didn't parsed the image data
#10946 opened Sep 26, 2025 by Xu-Wenqing Loading…
4 tasks
6
Fuse KV write + RoPE for Llama
#10945 opened Sep 26, 2025 by b8zhong Loading…
1 task done
4
[1/2] Support FA4 for MHA Prefill in sgl-kernel high priority run-ci
#10940 opened Sep 26, 2025 by lifuhuang Loading…
@zhyncs
3
[2/2] Support MHA prefill with FlashAttention 4. DO NOT MERGE high priority run-ci
#10937 opened Sep 26, 2025 by lifuhuang Loading…
4 tasks
@zhyncs
10
fix: make inference deterministic for large TP run-ci
#10930 opened Sep 26, 2025 by JustinTong0323 Loading…
4 tasks
1
5
ProTip! Mix and match filters to narrow down what you’re looking for.