Skip to content

Issues: sgl-project/sglang

Development Roadmap (2024 Q4)
#1487 opened Sep 21, 2024 by Ying1123
Open 22
[Feature] DeepSeek V3 optimization
#2591 opened Dec 26, 2024 by zhyncs
Open 18
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

193 Open 726 Closed
193 Open 726 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[Bug] Decode Throughput Inconsistency Between bench_serving and Engine Logs
#3050 opened Jan 22, 2025 by leepoly
5 tasks done
[Help wanted] CANN'T capture GPU activities using nsight system
#3049 opened Jan 22, 2025 by sleepwalker2017
[Feature] Reasoning model API support
#3043 opened Jan 22, 2025 by lambert0312
2 tasks done
[Bug] Qwen2-VL-7B with sglang Performance Degradation
#3041 opened Jan 22, 2025 by yileld
5 tasks done
[Feature]
#3040 opened Jan 22, 2025 by moxiegushi
2 tasks
[Feature] Support Beam Search enhancement New feature or request
#3032 opened Jan 21, 2025 by laixinn
2 of 4 tasks
@ispobock @HandH1998
Can router support --api-key parameter router
#3031 opened Jan 21, 2025 by lambert0312
@ByronHsu
2
[Feature] FP8 weight only w8a16 quantization native support quant LLM Quantization
#3007 opened Jan 20, 2025 by arunpatala
2 tasks done
@zhaochenyang20
1
what is the most efficient way to do with a 72b model and 8 * A100 ?
#3002 opened Jan 20, 2025 by Chandler-Bing
2
[Feature] Add docs for Offline Engine token-in token-out documentation Improvements or additions to documentation good first issue Good for newcomers RLHF Using SGLang for post training
#2968 opened Jan 18, 2025 by zhaochenyang20
2 tasks
@zhaochenyang20
2
[Feature] remove vllm _custom_ops good first issue Good for newcomers help wanted Extra attention is needed high priority
#2965 opened Jan 18, 2025 by zhyncs
7 tasks
@bjmsong @ByronHsu
@ispobock @BBuf @HandH1998 @zhyncs
1
QVQ Prefill stage slow
#2961 opened Jan 18, 2025 by WuNein
3
[Bug] Regex isn't precluding parentheticals. And maybe more. help wanted Extra attention is needed
#2957 opened Jan 17, 2025 by cinjon
5 tasks done
@zhaochenyang20
6
[Bug] JSONResponse fails if the probability distribution is very spiky.
#2955 opened Jan 17, 2025 by cinjon
5 tasks done
1
[Feature] Add docs for local accuracy tests documentation Improvements or additions to documentation good first issue Good for newcomers
#2953 opened Jan 17, 2025 by zhaochenyang20
2 tasks
@zhaochenyang20
9
[Feature] Enhancement on Sparse Attention and KV-Cache Compression
#2946 opened Jan 17, 2025 by shadowpa0327
2 tasks done
1
[Bug] Unrecognized keys in rope_scaling for 'rope_type'='yarn': {'original_max_position_embeddings'} help wanted Extra attention is needed
#2943 opened Jan 17, 2025 by rangehow
5 tasks done
@zhaochenyang20
1
[Feature] support EAGLE 2 with Triton Backend good first issue Good for newcomers help wanted Extra attention is needed high priority
#2940 opened Jan 17, 2025 by zhyncs
2 tasks
[Bug] KeyError: 'lm_head.weight' when loading quantized llama 3.2 3B and 1B models
#2935 opened Jan 17, 2025 by arunpatala
5 tasks done
5
[Bug] def get_nvgpu_memory_capacity() causes crash on NVIDIA H100 MIG
#2933 opened Jan 17, 2025 by dsingal0
5 tasks done
3
[Feature] (WIP)Support disaggregated serving to separate prefill and decoding
#2932 opened Jan 17, 2025 by zhaohaidao
2 tasks
[Bug] tensor_model_parallel_all_reduce' is not defined
#2931 opened Jan 17, 2025 by bakch92
2
[Feature] Lora optimization
#2929 opened Jan 16, 2025 by Fridge003
2 of 11 tasks
Warning while running Deepseek-V3 amd
#2921 opened Jan 16, 2025 by ishaandatta
@zhaochenyang20
1
[Bug] [OpenAI compatible API] Chunks of tokens aren't being split into separate indexes when specifying n > 1 generations documentation Improvements or additions to documentation
#2912 opened Jan 16, 2025 by accupham
5 tasks done
@zhaochenyang20
4
ProTip! Find all open issues with in progress development work with linked:pr.