Skip to content

Issues: vllm-project/vllm

[Roadmap] vLLM Roadmap Q1 2025
#11862 opened Jan 8, 2025 by simon-mo
Open 7
[V1] Feedback Thread
#12568 opened Jan 30, 2025 by simon-mo
Open 63
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

1,432 Open 5,710 Closed
1,432 Open 5,710 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]: Gemma3 Offline Batch Inference: Attempted to assign abc multimodal tokens to ab(c+2) placeholders bug Something isn't working
#14897 opened Mar 16, 2025 by BiEchi
1 task done
1
[Feature]: 使用guided_decoding来实现Function Calling feature request New feature or request
#14890 opened Mar 16, 2025 by shell-nlp
1 task done
[Usage]: ModuleNotFoundError: No module named 'triton' usage How to use vllm
#14888 opened Mar 16, 2025 by Hinsael
1 task done
[Bug]:0.74 dev ,the error occurred in the gptq_marlin_gemm function call bug Something isn't working
#14887 opened Mar 16, 2025 by su400
1 task done
[Bug]: RuntimeError: CUDA error: invalid argument bug Something isn't working
#14885 opened Mar 16, 2025 by vladlen32230
1 task done
[Bug]: if chat_template loaded from disk, jinja exception thrown from _try_extract_ast() bug Something isn't working
#14884 opened Mar 16, 2025 by hibukipanim
3
[Usage]: DeepSeek R1 input tokens cannot exceed 32k and how to correctly use FlashMLA usage How to use vllm
#14882 opened Mar 16, 2025 by FlintyLemming
1 task done
1
[New Model]: Command A with tool support new model Requests to new models
#14866 opened Mar 15, 2025 by Hexoplon
1 task done
[Usage]: What should I do if I want to skip the prefill of a new request? usage How to use vllm
#14863 opened Mar 15, 2025 by chenhongyu2048
1 task done
[Installation]: How to complete the installation of the latest vllm offline through code installation Installation problems
#14856 opened Mar 15, 2025 by Wandermay
1 task done
[Bug]: vLLM ModelConfig doesn't pass hf_overrides to get_hf_image_processor_config, which could contain auth token for hugging face (not in ENV) bug Something isn't working good first issue Good for newcomers
#14854 opened Mar 15, 2025 by void-mckenzie
1 task done
7
[Bug]: TTFT Performance Regression in vLLM v0.7.0 Compared to v0.6.1.post2 bug Something isn't working
#14845 opened Mar 14, 2025 by asleepykitty
1 task done
1
[Feature]: specify model only in config.yaml feature request New feature or request good first issue Good for newcomers
#14819 opened Mar 14, 2025 by g0t4
1 task done
1
1
[Usage]: Vllm whisper model response_format verbose_json not working usage How to use vllm
#14818 opened Mar 14, 2025 by deepakkumar07-debug
2
[Usage]: max_model_len, max_num_seqs and mm_counts usage How to use vllm
#14816 opened Mar 14, 2025 by TheFloHub
1
[Bug]: Gemma-3-27b-it-GPTQ Can't run in sm75, vllm-0.7.4.dev bug Something isn't working
#14814 opened Mar 14, 2025 by HelloCard
1 task done
2
[Installation]: uv run vllm serve "Qwen/Qwen2-VL-7B-Instruct" cannot start. uvloop does not support windows installation Installation problems
#14813 opened Mar 14, 2025 by metero20000
1 task done
3
[Usage]: how to reduce the number of processes of compile_worker usage How to use vllm
#14808 opened Mar 14, 2025 by FanYaning
1 task done
@youkaichao
3
[Bug]: CUDA_VISIBLE_DEVICES is not supported bug Something isn't working
#14807 opened Mar 14, 2025 by chenhongyu2048
1 task done
2
[Usage]: Are 4090 graphics cards of different brands and clock frequencies compatible, especially when using vLLM? usage How to use vllm
#14802 opened Mar 14, 2025 by xajhlhl
1 task done
[Bug]: v0.7.4 dev version CPU usage remains at 100% even when no requests are being processed. bug Something isn't working
#14799 opened Mar 14, 2025 by AndrewTsao
1 task done
[Usage]: How to make sure the timeout takes effect usage How to use vllm
#14792 opened Mar 14, 2025 by DayDayupupupup
1 task done
[Bug]: GLM4V model gets lower precision score on TextVQA since vLLM does not process model's position ids correctly. bug Something isn't working
#14790 opened Mar 14, 2025 by FrankiYYF
1 task done
1
[Usage]: How to add Sampling Parameters (n and best_of ) for benchmark_serving.py usage How to use vllm
#14775 opened Mar 13, 2025 by VinayHN1365466
1 task done
1
[V1][Bug] IMA with ngram spec decoding and flashinfer bug Something isn't working
#14765 opened Mar 13, 2025 by markmc
1 task done
1
ProTip! Adding no:label will show everything without a label.