Vulnerability details

CVE-2026-22778

vllm-project vLLM

vLLM leaks a heap address when PIL throws an error

CVSS 3.1
9.8Critical

Official 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.

CVSS scores

ScoreVersionSourceVector
9.8Critical 3.1 CVEProject cvelistV5 (CNA) (GitHub_M) CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
9.8Critical 3.1 CVEProject cvelistV5 (CNA) (redhat-SADP) CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

Affected products

VendorProductVersionsSource
vllm-project vLLM >= 0.8.3, < 0.14.1 CVEProject cvelistV5 (CNA)

References and advisories

15
Kolun on Telegram @kolunsec Exploited CVEs, key news and a daily digest Alerts for this product's next CVEsFollow the product to get new vulnerabilities on Telegram.

Details

Published
2026-02-03
Last updated
2026-07-15
State
Published
Assigner (CNA)
GitHub_M

Exploit prediction (EPSS)

18.28%
Probability of exploitation in 30 days
97.2%
Percentile (among all CVEs)

SSVC assessment (CISA)

Automatable
Yes
Exploitation
None
Technical impact
Total

Weaknesses (CWE)

  • CWE-532
    Insertion of Sensitive Information into Log File
  • CWE-209
    Generation of Error Message Containing Sensitive Information