vllm has 106 CVEs on record between 2024 and 2026. Disclosure cadence is accelerating: 54 in the last 90 days against 15 in the 90 before. The busiest recent month was September 2026 with 25. The median CVSS is 6.5 (medium), with 10 rated critical. None have a confirmed exploitation report. The dominant weakness classes are CWE-770 (22) and CWE-400 (14).

CVEs per month

Weakness classes

Products

Worst active — by depth score

vllm vulnerabilities

vLLM Tool Schema allows DoS via Malformed pattern and type Fields

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Remote Code Execution Vulnerability in vLLM Multi-Node Cluster Configuration

vLLM Vulnerable to Remote Code Execution via Mooncake Integration

phi4mm: Quadratic Time Complexity in Input Token Processing​ leads to denial of service

Data exposure via ZeroMQ on multi-node vLLM deployment

CVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0

Rejected reason: This CVE ID has been rejected or withdrawn by its CVE Numbering Authority.

vLLM allows Remote Code Execution by Pickle Deserialization via AsyncEngineRPCServer() RPC server entrypoints

vLLM Deserialization of Untrusted Data vulnerability

vLLM Allows Remote Code Execution via Mooncake Integration

vLLM denial of service via outlines unbounded cache on disk

vLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cache

vllm: Malicious model to RCE by torch.load in hf_model_weights_iterator

vLLM Denial of Service via the best_of parameter

vLLM denial of service vulnerability

phi4mm: Quadratic Time Complexity in Input Token Processing leads to denial of service