---
id: CVE-2026-22773
title: vLLM is an inference and serving engine for large language models (LLMs)
summary: >-
  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 s…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-770
vendor: vllm
product: vllm
affected:
  - 'vllm >= 0.6.4, < 0.12.0'
patched:
  - vllm 0.12.0
published: '2026-01-10'
updated: '2026-06-17'
sourceUpdated: '2026-06-17T10:20:24.027'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-22773'
references:
  - url: >-
      https://github.com/vllm-project/vllm/security/advisories/GHSA-grg2-63fw-f2qr
    label: security-advisories@github.com
tags:
  - nvd
  - cve.org
ssvc:
  exploitation: none
  automatable: 'no'
  technicalImpact: partial
  timestamp: '2026-01-12T13:22:42.362326Z'
ingestedAt: '2026-09-13T10:28:38.840Z'
epss: 0.00449
epssPercentile: 0.36259
---

## Overview

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.

## Affected

- `vllm >= 0.6.4, < 0.12.0`

## Remediation

Upgrade past the affected range:

- `vllm 0.12.0`