Information disclosure in Vllm

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.

Published · last modified .

CVSS v3 metric

CVSS v3 base score 9.8 (Critical). Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H.

EPSS exploit prediction

EPSS: 0.167 (97.0th percentile), scored .

Moderate probability of exploitation in the next 30 days — watch for active exploitation reports. 97th percentile — 97.0% of CVEs in the catalogue have a lower EPSS than this one. How to read EPSS.

EPSS trend (30 days)EPSS over the last 30 days for CVE-2026-22778: rose from 0.037 to 0.167.

EPSS over last 30 days · oldest: 0.037 · newest: 0.167 · change: +0.130

Affected products

Weakness classification (CWE)

Public proof-of-concept exploits

References

Frequently asked questions

What is CVE-2026-22778?
CVE-2026-22778 is a critical-severity vulnerability in Vllm, classified under Insertion of Sensitive Information into Log File. CVSS score: 9.8/10. Published 2026-02-02.
How severe is CVE-2026-22778?
Critical severity. CVSS v3 base score is 9.8 out of 10.
Is CVE-2026-22778 known to be exploited?
2 public proof-of-concept repositories are indexed. Not currently listed in the CISA KEV catalog.