VULNERABILITY INTELLIGENCE

CVE-2026-22773

CVSS score6.5 MEDIUM
EPSS probability0.44%
CISA KEVNot currently listed
Published2026-01-10
Last modified2026-06-17
CVSS vectorCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H

Description

vLLM is an inference and serving engine for large language models (LLMs). In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimension mismatch that results in an unhandled runtime error, leading to complete server termination. This issue has been patched in version 0.12.0.

Weakness classification

  • security-advisories@github.com: CWE-770

Primary references