CVE-2026-41523: vLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code Execution

Severity: high (CVSS 7.5 v3.1)
Exploited in the wild: not listed in CISA KEV
EPSS: 0.9% probability of exploitation in the next 30 days
Public exploit or detection template: yes
Published: 2026-06-22 · updated 2026-10-25
Weakness: CWE-94

Description

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1). This vulnerability is fixed in 0.22.0.

Affected products

vllm-project vllm: before 0.22.0 — fixed in 0.22.0

How to fix

Upgrade to a fixed version: vllm-project vllm 0.22.0. Check the vendor advisory for mitigations.

References

Get alerted when CVE-2026-41523 or the next vulnerability affects your stack