vllm vulnerabilities (PyPI)
113 vulnerabilities.
vllmPyPICVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.09.8 CriticalvllmPyPIvLLM vulnerable to Denial of Service by abusing xgrammar cache6.5 MediumvllmPyPIRemote Code Execution in vllm-project/vllm9.8 CriticalvllmPyPIRemote Code Execution in vllm-project/vllm9.8 CriticalvllmPyPIvLLM deserialization vulnerability in vllm.distributed.GroupCoordinator.recv_object9.8 CriticalvllmPyPIvLLM Allows Remote Code Execution via Mooncake Integration9.1 CriticalvllmPyPIvLLM denial of service via outlines unbounded cache on disk6.5 MediumvllmPyPIvLLM using built-in hash() from Python 3.12 leads to predictable hash collisions in vLLM prefix cache2.6 LowvllmPyPIvLLM allows a malicious model RCE by torch.load in hf_model_weights_iterator7.5 HighvllmPyPIVllm: denials of service in vllm json web api6.2 MediumvllmPyPIVllm: a completions api request with an empty prompt will crash the vllm api server.7.5 High