Skip to content

Issues: sgl-project/sglang

Development Roadmap (2024 Q4)
#1487 opened Sep 21, 2024 by Ying1123
Open 26
[Feature] DeepSeek V3 optimization
#2591 opened Dec 26, 2024 by zhyncs
Open 39
[Track] DeepSeek V3/R1 nextn progress
#3472 opened Feb 10, 2025 by zhyncs
Open
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

260 Open 845 Closed
260 Open 845 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

Hang at around 4% during the CUDA graph loading process
#3547 opened Feb 13, 2025 by BoBo0037
@minleminzui
1
[Feature] Support multimodal models for Native API/ Engine feature visIon-LM
#3545 opened Feb 13, 2025 by shuaills
2 tasks done
@minleminzui
[Bug] Multiple node deployment of deepseek-v3 reported an error. Regarding NCCL, there were no environment settings for nccl when installing the sglang environment
#3544 opened Feb 13, 2025 by XIEDEI
5 tasks done
@minleminzui
[Feature] Request to Support Additional Embedding Models (e.g., DMetaSoul/Dmeta-embedding-zh)
#3539 opened Feb 13, 2025 by q77889966
2 tasks done
@minleminzui
1
[Bug]NCCL error if enable the cuda graph bug Something isn't working high priority
#3538 opened Feb 13, 2025 by LJL36
5 tasks done
@ispobock @zhyncs @minleminzui
1
[Feature] Respect max_completion_tokens feature
#3531 opened Feb 12, 2025 by siddartha-RE
2 tasks done
@jhinpan
2
[Feature] Extend CustomLogitProcessor to Support input_ids in call Method feature
#3524 opened Feb 12, 2025 by Aykhan-sh
1 of 2 tasks
@jhinpan
2
[Bug] sglang doesn't stop the generation when the request is canceled enhancement New feature or request
#3520 opened Feb 12, 2025 by tchaton
5 tasks done
@jhinpan
1
[Bug] maximum recursion depth exceeded
#3518 opened Feb 12, 2025 by kebe7jun
2 of 5 tasks
1
@jhinpan
4
i use 3 A800 to deploy deepseek r1，but one A800 just one IB，how i adjust the number of tp in the deploy command deepseek
#3517 opened Feb 12, 2025 by 19157681683
@jhinpan
1
[Bug] use Eagle with llama3-8B
#3513 opened Feb 12, 2025 by feifeibear
5 tasks
@jhinpan
1
something wrong when use SLURM for Multi-Node Inference
#3512 opened Feb 12, 2025 by Tian14267
@jhinpan
6
Proper way to stop nsys profiling for SGLang serving help wanted Extra attention is needed
#3511 opened Feb 12, 2025 by chenzhengda
@jhinpan
2
[Bug] Get error when use two nodes
#3510 opened Feb 12, 2025 by Tian14267
5 tasks
@jhinpan
2
Why is the throughput of A40 15 token/s, slower than vllm?
#3508 opened Feb 12, 2025 by EvanSong77
@jhinpan
1
In distributed scenarios, how to configure k8s ports and which ports need to be opened?
#3489 opened Feb 11, 2025 by markluofd
@shuaills
5
[Bug] It will stuck when run python -m sglang.launch_server
#3488 opened Feb 11, 2025 by Tian14267
5 tasks
@shuaills
7
[Question] How Sglang balances workload?
#3487 opened Feb 11, 2025 by vabatista
@shuaills
1
[Track] DeepSeek V3/R1 accuracy
#3486 opened Feb 11, 2025 by zhyncs
1
[Feature] is flex attention compatible? amd help wanted Extra attention is needed
#3479 opened Feb 11, 2025 by johnnynunez
2 tasks done
@zhaochenyang20
1
deepseek-R1 AssertionError occurred in the batch request of the client deepseek
#3477 opened Feb 11, 2025 by Roysky
@zhaochenyang20
2
[Track] DeepSeek V3/R1 nextn progress deepseek enhancement New feature or request flashinfer high priority
#3472 opened Feb 10, 2025 by zhyncs
8 of 13 tasks
@zhyncs
[Track] long context performance sglang vs vllm flashinfer high priority performance
#3471 opened Feb 10, 2025 by zhyncs
@zhyncs
2
Can not load Deepseek model in two nodes
#3467 opened Feb 10, 2025 by Tian14267
@zhaochenyang20
8
[Bug] Qwen2 Eagle serving error high priority
#3465 opened Feb 10, 2025 by feifeibear
5 tasks
@zhyncs
3
ProTip! Follow long discussions with comments:>50.