← Back

CVE-2026-22778

Published: Feb 2, 2026Modified: Jul 15, 2026

JSON object

Loading...
9.8
Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Exploitability: 3.9 / Impact: 5.9
Source: security-advisories@github.com (Secondary)

Description

vLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1.

Affected (1)

Products: Vllm: Vllm
1 product
Vllm
Configuration A
1 vulnerable
Vulnerable SoftwareAffected Versions
From 0.8.3 to 0.14.1

References (15)

Source: security-advisories@github.com
Issue TrackingPatch
Source: security-advisories@github.com
Issue TrackingPatch
Source: security-advisories@github.com
Release Notes
Source: security-advisories@github.com
Vendor Advisory
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c
Source: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c

Timeline

No history available yet.