-
-
Notifications
You must be signed in to change notification settings - Fork 5.9k
Issues: vllm-project/vllm
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[Usage]: Speculative Decoding KV Cache Generate
usage
How to use vllm
#13845
opened Feb 25, 2025 by
qingyiaaaaa
1 task done
[Feature]: Add CLI Commands for Benchmarking
feature request
good first issue
Good for newcomers
#13840
opened Feb 25, 2025 by
simon-mo
1 task done
[Bug]: database disk image is malformed
bug
Something isn't working
#13838
opened Feb 25, 2025 by
abpani
1 task done
[Bug]: restarting vllm --> "WARNING: process group has NOT been destroyed before we destruct ProcessGroupNCCL"
bug
Something isn't working
#13836
opened Feb 25, 2025 by
AlbiRadtke
2 tasks done
[Feature]: Support triton 3.3.0
feature request
#13829
opened Feb 25, 2025 by
johnnynunez
1 task done
[Bug]: Gibberish Output from LLaMA 3.1 8B using vLLM with xGrammar
bug
Something isn't working
#13828
opened Feb 25, 2025 by
UTSAV-44
1 task done
[Feature]: Support for ColBERT (Late-Interaction Retrieval) in vLLM
feature request
#13827
opened Feb 25, 2025 by
FernandoDorado
[Bug]: vllm0.7.3: an illegal memory access was encountered
bug
Something isn't working
#13824
opened Feb 25, 2025 by
tingjun-cs
1 task done
[Bug]: Structured generation with JSON schema does not produce empty array
bug
Something isn't working
#13821
opened Feb 25, 2025 by
Nithin-Holla
1 task done
[Bug]: After updating VLLM from 0.4.0post1 to the 0.7.3, the model cannot load properly
bug
Something isn't working
#13819
opened Feb 25, 2025 by
DominikZhuan
1 task done
[Bug]: Qwen2.5-VL-3B Returning Series of !!! for Specific Image when "dtype" Set to "float16"
bug
Something isn't working
#13817
opened Feb 25, 2025 by
MotorBottle
1 task done
[Performance]: Regarding PD separation performance
performance
Performance-related issues
#13816
opened Feb 25, 2025 by
skyCreateXian
1 task done
[Bug]: ValueError: invalid literal for int() with base 10 for device_id_to_physical_device_id` function
bug
Something isn't working
#13815
opened Feb 25, 2025 by
googlercolin
1 task done
[Bug]: vLLM serve Deepseek-R1-14B tensor parallelism failed
bug
Something isn't working
#13814
opened Feb 25, 2025 by
caolonghao
1 task done
[Bug]: vLLM serve Deepseek-R1 on 4x8*L20 cluster failed
bug
Something isn't working
#13812
opened Feb 25, 2025 by
ly01325
1 task done
[Usage]: vllm v0.7.2 can not support baichuan2 model
usage
How to use vllm
#13810
opened Feb 25, 2025 by
alanayu
1 task done
[Bug]: using v1 AsyncLLMEngine, signal only works in main thread of the main interpreter. v0 has no this problem.
bug
Something isn't working
#13806
opened Feb 25, 2025 by
Superhors
1 task done
[Bug]: The accuracy of multiple cards and single card is inconsistent
bug
Something isn't working
#13801
opened Feb 25, 2025 by
jiangix-paper
1 task done
[Feature]: [V1] Parallel sampling should support metrics
feature request
#13792
opened Feb 24, 2025 by
afeldman-nm
1 task done
[Bug]: ValueError: Unsupported FA version: None on V100 and V1 engine
bug
Something isn't working
#13788
opened Feb 24, 2025 by
piotr-sikora-v
1 task done
[Bug]: vllm server hang when running DeepSeek R1
bug
Something isn't working
#13778
opened Feb 24, 2025 by
yangsijia-serena
1 task done
[Misc] [ROCm]: Build from source failure with Arch/gcc14 with ROCm 6.3
misc
#13777
opened Feb 24, 2025 by
arjunkathuria
1 task done
[Usage]: Cannot use FA version 2 is not supported due to FA3 is only supported on devices with compute capability >= 8 excluding 8.6 and 8.9
usage
How to use vllm
#13766
opened Feb 24, 2025 by
Like-Drinking-water
1 task done
Previous Next
ProTip!
What’s not been updated in a month: updated:<2025-01-25.