A AegiFlow
HIGHCVSS 8.8EPSS 0.7%

CVE-2026-22807

CVE-2026-22807 updated by NVD

Published
2026-01-21
Modified
2026-07-23
EPSS percentile
51%
Sources
github-advisory, nvd

Summary

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.

Affected packages

EcosystemPackageAffected versionsFixed versions
PyPIvllm0.14.0

Remediation: Upgrade to 0.14.0 or later.

References

Includes data from the GitHub Advisory Database, licensed under CC-BY 4.0.

Includes data from the National Vulnerability Database (NIST). NVD data is in the public domain; this page is not endorsed by NIST.

CVE® is a registered trademark of The MITRE Corporation. CVE content reproduced under the CVE Terms of Use; copyright designation © MITRE.

EPSS scores provided by the FIRST.org Exploit Prediction Scoring System.