CVE-2026-22807

CVSS 8.8 HighEPSS 0.0%
  • CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
  • Published: 2026-01-21T22:15:49.077

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.

Related CVE by CWE

No related CWE found.

Top CVE for Vendor

No vendor taxonomy on this entry.

Recently Exploited Similar Vulnerabilities

No recent KEV-listed items for this vendor/product.

How to fix CVE-2026-22807

CVE-2026-22807 is a high severity vulnerability affecting the affected product.

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 […]

Exploit Difficulty: MEDIUM
⏱️ Time to exploit: 1-4 hours
🛠️ Required skills: Intermediate security knowledge
💰 Public exploits: May be available

How to Fix:

1 Identify affected systems

- Check if you're running the affected product

2 Immediate actions

- Update to the latest patched version
- If patching is not immediately possible: restrict network exposure, apply least-privilege access

3 Verification

- Test the fix in a staging environment first
- Review logs for signs of exploitation
- Monitor for IOCs (Indicators of Compromise)

4 Long-term prevention

- Enable automatic security updates
- Set up vulnerability monitoring
- Review and harden security configurations

Exploit Difficulty Assessment

MEDIUM
⏱️ Time to Exploit: 1-4 hours
🛠️ Skills Required: Intermediate security knowledge
💰 Public Exploits: May be available

Vulnerability Timeline

Jan 21, 2026
Vulnerability Published

CVE details first published to NVD database

Jan 21, 2026
Imported to Database

Added to this CVE tracking system

Detection Rules & IOCs

No specific detection rules generated for this vulnerability type.

No vendor/product data available.