Vulnerability Analysis
vLLM leaks a heap address when PIL throws an error
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.
Nuclei Template
CVSS Base Score
9.8
CRITICAL
Exploitability:3.9
Impact Score:5.9
Temporal Score:-
EPSS:3.82%
Threat Intelligence Signals
CISA KEV
No
KEV Date Added
—
Ransomware Use
—
KEV Due Date
—
VulnCheck In-the-Wild
No
Nuclei Template
YES
EPSS Score
3.816%
EPSS Percentile
88.8th pct
GHSA ID
GitHub Severity
CRITICAL
SSVC Exploitation
None
SSVC Automatable
Yes
Vulnerability Class
—
Identity & Timeline
| Status | - |
| Assigning Authority | - |
| Discovered Via | - |
| Record Completeness | - |
| CVSS Version / Source | - |
| Reserved | - |
| Published | - |
| Patch Date (date_public) | - |
| Exploit DB Date | - |
| First GitHub PoC Date | - |
| Last Updated | - |
| Time to Patch (Days to fix) | - |
| Exploit Release Gap | - |
| PoC Release Gap | - |
| Exploit DB References | None identified |
Affected Products & Versions
| Vendor | Product | Affected Versions |
|---|---|---|
| No affected products specified. | ||
Social Buzz