[CVE-2026-41523] vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert…

Descripción técnica

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.

Puntuación CVSS

Score: 7.5/10 — Severidad: HIGH — Estado NIST: Modified

CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H

Productos afectados

Vllm — Vllm

Debilidades (CWE)

CWE-94, CWE-617

Fuente oficial

Publicado en NIST NVD.