CVE-2026-22807
Last modified
CVE-2026-22807 is a critical-severity vulnerability rated 9.8/10 on the CVSS scale. 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. EPSS estimates a 0.54% chance of exploitation in the next 30 days.
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.
Metrics
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Weakness Enumeration
Affected Software
| Vendor | Product | Versions |
|---|---|---|
| Vllm | Vllm | >= 0.10.1, < 0.14.0 |
References
- https://github.com/vllm-project/vllm/pull/32194Issue Tracking, Patch
- https://github.com/vllm-project/vllm/releases/tag/v0.14.0Product, Release Notes
- https://github.com/vllm-project/vllm/security/advisories/GHSA-2pc9-4j83-qjmrPatch, Vendor Advisory
Timeline
- Published
- Last Modified
- Status
- Analyzed
Frequently Asked Questions
What is CVE-2026-22807?
How severe is CVE-2026-22807?
How do I fix CVE-2026-22807?
Are you affected by CVE-2026-22807?
Run a free Strix scan to check your systems for this vulnerability.
Scan your code nowSource: NVD / NIST