CVE-2026-22807PYSEC-2026-2010GHSA-2PC9-4J83-QJMR
vLLM affected by RCE via auto_map dynamic module loading during model initialization
- Published
- Last modified
- Where the record stands: published, rejected by its publisher, or superseded by another id.What do these mean? →
- published
- Whether CISA lists it in its Known Exploited Vulnerabilities catalog: proof that attackers are using it.What do these mean? →
- Not listed in CISA KEV
- Fix
- Fixed in 0.14.0
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- 1 proof of concept on GitHub
CISA's decision points: whether it is exploited, whether attacks can be automated, and how much control an attacker gets. Assessed (SSVC 2.0.3).What do these mean? →
- Exploitation: none
- Automatable: no
- Technical impact: total
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.
Timeline
- CVE published
- CISA SSVC assessment
Exploitation: none · Automatable: no · Technical impact: total
- Record last updated
Affected versions
- vllm-project vllm
From 0.10.1 · fixed in 0.14.0
- vllmPyPI
From 0.10.1 · fixed in 0.14.0