Skip to content
CVE PROGRAM SOURCE RECORD9.8 Critical

CVE-2026-22778

vLLM leaks a heap address when PIL throws an error

Read JSON ↗CVE.org ↗
Status
Published
Published
Updated
Assigning organization
GitHub_M

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 products

Version statuses are supplied by the assigning organization. Consult the original record for detailed ranges and platform conditions.

vllm-project vllm

  • >= 0.8.3, < 0.14.1affected

References