Vulnerability Report

CVE-2026-22778

RCE

Title: Heap Address Leak in vLLM

Memory Corruption

Proof Of Concept

PoC Available for CVE-2026-22778

CWE Category CWE-532

Published Date Feb 02, 2026

Modified Date Jul 15, 2026

Exploit Status Available

Score 9.8 CVSS v3.1

Exploit Probability (EPSS)

3.80%

Vulnerability Summary

CVE-2026-22778: 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.

Impacted Vendors

vllm 1

vllm-project 1

Reference Links

https://github.com/vllm-project/vllm/pull/31987 https://github.com/vllm-project/vllm/pull/32319 https://github.com/vllm-project/vllm/releases/tag/v0.14.1 https://github.com/vllm-project/vllm/security/advisories/GHSA-4r2x-xpjr-7cvv

CVSS v3.1

Source Entity [email protected]

Severity CRITICAL

9.8

Attack Vector

NETWORK

Complexity

LOW

Privileges

N/A

Interaction

NONE

Confidentiality

N/A

Integrity

N/A

Availability

N/A

Scope

UNCHANGED

RAW VECTOR CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H