Menu

Star on GitHub · 3.2k

CVE-2026-22807

8.8 · HIGH

vLLM is an inference and serving engine for large language models (LLMs).

Published Jan 21, 2026CWE-94EPSS 0.82%

Overview

CVE-2026-22807 is a high-severity vulnerability. It was published on January 21, 2026 and has a CVSS 3.1 base score of 8.8 (HIGH).

This vulnerability has a CVSS 3.1 base score of 8.8, rated HIGH. It can be exploited remotely over the network. No authentication or special privileges are required for exploitation.

Technical 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.

Remediation

Check the references section for vendor advisories, patches, and mitigation guidance. If immediate patching is not possible, review the CVSS vector to understand the attack surface and apply compensating controls such as network segmentation or access restrictions.

Frequently asked questions

What is CVE-2026-22807?

CVE-2026-22807 is a high-severity vulnerability. It was published on January 21, 2026 and has a CVSS 3.1 base score of 8.8 (HIGH).

How severe is CVE-2026-22807?

This vulnerability has a CVSS 3.1 base score of 8.8, rated HIGH. It can be exploited remotely over the network. No authentication or special privileges are required for exploitation.

How do I fix or remediate CVE-2026-22807?

Check the references section for vendor advisories, patches, and mitigation guidance. If immediate patching is not possible, review the CVSS vector to understand the attack surface and apply compensating controls such as network segmentation or access restrictions.

How can CyberStrike help with CVE-2026-22807?

CyberStrike is an open-source, AI-augmented pentesting engine. Rather than flag a version number, it tests whether an issue like CVE-2026-22807 is actually reachable in your own applications and APIs, then verifies each finding before reporting it. That tells you if the weakness is a real, exploitable risk in your environment or a false positive you can safely deprioritize.

Related HIGH CVEs from 2026

View all