Skip to content

Pull requests: sgl-project/sglang

New pull request New
217 Open 3,039 Closed
217 Open 3,039 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

fix multimodal hash feature
#5083 opened Apr 5, 2025 by huangtingwei9988 Loading…
1 of 8 tasks
2
Fix prefill OOM error in the case of large page size bug Something isn't working
#5081 opened Apr 5, 2025 by xiezhq-hermann Loading…
1 of 6 tasks
@merrymercy
python transfer custom allreduce from trt kernel to vllm kernel
#5080 opened Apr 5, 2025 by yizhang2077 Loading…
6 tasks
sgl-kernel transfer custom allreduce from trt kernel to vllm kernel high priority
#5079 opened Apr 5, 2025 by yizhang2077 Loading…
6 tasks
[Model] Support ArcticForCausalLM architecture (Snowflake/snowflake-arctic-instruct)
#5078 opened Apr 5, 2025 by b8zhong Loading…
2 tasks done
1
Tiny extract launch_server_process
#5076 opened Apr 5, 2025 by fzyzcjy Loading…
6 tasks
Support benchmarking with presets on multiple configuration combinations
#5075 opened Apr 5, 2025 by fzyzcjy Loading…
6 tasks
[Model] Add support for nvidia/Llama-3_3-Nemotron-Super-49B-v1
#5073 opened Apr 4, 2025 by kylehh Loading…
6 tasks
update sgl-kernel build scripts for CUDA compatibility
#5071 opened Apr 4, 2025 by fecet Loading…
6 tasks done
37
[fix] fix Qwen2ForSequenceClassification test help wanted Extra attention is needed
#5070 opened Apr 4, 2025 by Alcanderian • Draft
2 of 6 tasks
3
[Fix] DeepEP Compatibility with Low Latency high priority
#5068 opened Apr 4, 2025 by liz-badada • Draft
1 of 6 tasks
@ch-wan
2
Bump flashinfer to 0.2.5
#5067 opened Apr 4, 2025 by AkazaAkane Loading…
2 of 6 tasks
6
[VLM] Adopt fast image processor by default
#5065 opened Apr 4, 2025 by mickqian Loading…
6 tasks
3
[Model]support nvidia/Llama-3_3-Nemotron-Super-49B-v1
#5063 opened Apr 4, 2025 by lambda7xx Loading…
6 tasks
[Feat] platform auto detection high priority
#5059 opened Apr 4, 2025 by Alcanderian Loading…
1 of 6 tasks
add function call parser for DeepSeek V3
#5054 opened Apr 4, 2025 by finger92 Loading…
6 tasks
[WIP] adapt trtllm fp blockscale kernel
#5045 opened Apr 3, 2025 by yizhang2077 Loading…
6 tasks
fix: remove duplicate compressed-tensors in pyproject
#5044 opened Apr 3, 2025 by fecet Loading…
6 tasks done
needed imports to support vllm's compressed_tensors quantized models
#5042 opened Apr 3, 2025 by Sadeghi85 Loading…
fix gemma 3 error when config.json is set to text but model is multimodal
#5041 opened Apr 3, 2025 by Sadeghi85 Loading…
Support BNB quantization for llama/mllama
#5038 opened Apr 3, 2025 by ryang-max Loading…
4 of 7 tasks
Tiny update error hint
#5037 opened Apr 3, 2025 by fzyzcjy Loading…
6 tasks
Add SSL options to launch_server
#5033 opened Apr 3, 2025 by k4rth33k Loading…
3 of 6 tasks
1
Fix test_flashattn_backend circular import and param missing bugs
#5028 opened Apr 3, 2025 by WhatGhost Loading…
1 of 6 tasks
fix(sampler): logprobs overflow due to non-deterministic sort when using top_p sampling
#5027 opened Apr 3, 2025 by zhc7 Loading…
6 tasks done
1
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.