CVE-2026-22807

Severity CVSS v4.0:

Pending analysis

Type:

CWE-94 Code Injection

Publication date:

21/01/2026

Last modified:

21/01/2026

Description

vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face `auto_map` dynamic modules during model resolution without gating on `trust_remote_code`, allowing attacker-controlled Python code in a model repo/path to execute at server startup. An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load. This happens before any request handling and does not require API access. Version 0.14.0 fixes the issue.

Impact

Vector 3.x

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H CVSS v3.1 Severity and Metrics:

Base Score: 8.80 HIGH
Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Attack Vector (AV): Network
Attack Complexity (AC): Low
Privileges Required (PR): None
User Interaction (UI): Required
Scope (S): Unchanged
Confidentiality (C): High
Integrity (I): High
Availability (A): High

Base Score 3.x

8.80

Severity 3.x

HIGH

CVE-2026-22807

Description

Impact

References to Advisories, Solutions, and Tools