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.
CVSS Metrics
- Vector
- CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
- Attack Vector
- network
- Complexity
- low
- Privileges
- none
- User Action
- required
- Scope
- unchanged
- Confidentiality
- high
- Integrity
- high
- Availability
- high
- Weaknesses
- CWE-94
Metadata
- Primary Vendor
- VLLM
- Published
- 1/21/2026
- Last Modified
- 1/30/2026
- Source
- NIST NVD
- Note: Verify all details with official vendor sources before applying patches.
Affected Products
vllm : vllm
AI-Powered Remediation
Generate remediation guidance or a C-suite brief for this vulnerability.