Skip to content

Issues: vllm-project/vllm

[Roadmap] vLLM Roadmap Q1 2025
#11862 opened Jan 8, 2025 by simon-mo
Open 9
[V1] Feedback Thread
#12568 opened Jan 30, 2025 by simon-mo
Open 81
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

1,548 Open 5,977 Closed
1,548 Open 5,977 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

[Feature]: Support Cascade Attention for Sliding Window Attention feature request New feature or request good first issue Good for newcomers help wanted Extra attention is needed
#15710 opened Mar 28, 2025 by WoosukKwon
1 task done
[Bug]: docker scluster has no open port bug Something isn't working
#15707 opened Mar 28, 2025 by jli113
1 task done
1
[Bug]: Qwen2.5: Sliding window for some but all layers is not supported. This model uses sliding window but max_window_layers = 28 is less than num_hidden_layers = 28. Please open an issue to discuss this feature. bug Something isn't working
#15705 opened Mar 28, 2025 by Martin7-1
1 task done
@WoosukKwon
[Bug]: Engine crash periodically running Deepseek V3/R1 on Hopper GPUs in cutlass_scaled_mm_sm90() bug Something isn't working
#15702 opened Mar 28, 2025 by rymc
[Bug]: TP16 Not Working on k8s Cluster bug Something isn't working
#15698 opened Mar 28, 2025 by yongho-chang
1 task done
[Feature]: Composite model loading using AutoWeightsLoader for all models feature request New feature or request good first issue Good for newcomers
#15697 opened Mar 28, 2025 by DarkLight1337
1 task done
@lengrongfu
5
[Feature]: MiniCPM-O support beam_search feature request New feature or request
#15694 opened Mar 28, 2025 by yangJirui
1 task done
1
[Bug]: Performance downgrade 20% for AWQ + MLA + chunked-prefill bug Something isn't working
#15689 opened Mar 28, 2025 by DefTruth
1 task done
2
[Bug]: Worker died during distributed inference bug Something isn't working
#15687 opened Mar 28, 2025 by Haoxi2002
1 task done
[Bug]: ValueError: The checkpoint you are trying to load has model type qwen2_5_omni but Transformers does not recognize this architecture. bug Something isn't working
#15685 opened Mar 28, 2025 by jieguolove
1 task done
1
[Usage]: how to run two models in Docker usage How to use vllm
#15682 opened Mar 28, 2025 by Tu1231
1 task done
[Bug]: Qwen-2-vl inference question about max_seq_length bug Something isn't working
#15680 opened Mar 28, 2025 by Eduiskss
1 task done
1
[Feature]: K-cache only feature request New feature or request
#15679 opened Mar 28, 2025 by HaiFengZeng
1 task done
[Bug]: v0.8.2 generates incomplete sequences for Qwen2-VL-7B under specific concurrency range bug Something isn't working
#15677 opened Mar 28, 2025 by xushangning
1 task done
8
[Bug]: OpenAI-Compatible Server cannot be requested bug Something isn't working
#15675 opened Mar 28, 2025 by lihao056
1 task done
3
[Usage]: torchrun data parallel and tensor parallel at the same time usage How to use vllm
#15672 opened Mar 28, 2025 by JieFeng-cse
1 task done
[Feature]: update to flashinfer 0.2.3 feature request New feature or request
#15666 opened Mar 28, 2025 by youkaichao
1 task done
[Bug]: VLLM 0.8.2 OOM error (No error in 0.7.3 version) bug Something isn't working
#15664 opened Mar 28, 2025 by manitadayon
1 task done
6
[Bug]: v0.8.2 The default parameters start with a very low performance of 9.8 tokens/s with deepseek-r1:671b bug Something isn't working
#15663 opened Mar 28, 2025 by Hugh-yw
1 task done
2
[Feature]: distribute the package on macos feature request New feature or request
#15661 opened Mar 28, 2025 by youkaichao
1 task done
3
[Usage]: different models in the /vllm/model directory usage How to use vllm
#15660 opened Mar 28, 2025 by jeerychao
1 task done
[Feature]: Refactor the logic in tool parser manager and reasoning parser manager feature request New feature or request
#15658 opened Mar 28, 2025 by gaocegege
1 task done
[Bug]: vllm server dead issue bug Something isn't working
#15653 opened Mar 27, 2025 by pimang62
1 task done
1
[Performance]: Update Cascade Attention Heuristics for FA3 feature request New feature or request good first issue Good for newcomers help wanted Extra attention is needed
#15647 opened Mar 27, 2025 by WoosukKwon
1 task done
1
@lengrongfu
2
[Usage]: profiling.py example broken on v0.8.1 usage How to use vllm
#15640 opened Mar 27, 2025 by rajesh-s
1 task done
2
ProTip! Adding no:label will show everything without a label.