Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:HSummary
CVE-2026-22778 is a critical-severity Insertion of Sensitive Information into Log File (CWE-532) vulnerability in Vllm Vllm. Its CVSS base score is 9.8 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Unsecured Credentials (T1552); ranked in the top 11% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as NLP and Transformers; in the Privacy and Disclosure risk domain.
The strongest mitigations our analysis identified map to SI-11 (Error Handling) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2026-22778 is a heap address disclosure vulnerability (CWE-532) in vLLM, an open-source inference and serving engine for large language models (LLMs). The issue affects versions from 0.8.3 up to but not including 0.14.1. Specifically, when an invalid image is submitted to vLLM's multimodal endpoint, the Python Imaging Library (PIL) generates an error that vLLM echoes back to the client, inadvertently leaking a heap memory address. This disclosure has a CVSS v3.1 base score of 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H), reflecting its critical severity.
Unauthenticated remote attackers with network access to a vulnerable vLLM instance can exploit this by sending a crafted invalid image to the multimodal endpoint, obtaining a heap address leak that significantly weakens Address Space Layout Randomization (ASLR), reducing brute-force guesses from approximately 4 billion to around 8. This information disclosure can be chained with a separate heap overflow vulnerability in the JPEG2000 decoder implemented via OpenCV or FFmpeg, enabling full remote code execution (RCE) on the server.
The vulnerability is addressed in vLLM version 0.14.1, as detailed in the project's security advisory (GHSA-4r2x-xpjr-7cvv) and associated pull requests (#31987 and #32319). Security practitioners should upgrade to 0.14.1 or later and review exposure of multimodal endpoints, particularly in LLM serving environments. The release notes for v0.14.1 confirm the fix.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-5277
Vulnerability Data
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…
more
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.
- CWE(s)
AI Security AnalysisAI
- AI Category
- NLP and Transformers
- Risk Domain
- Privacy and Disclosure
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: llms, opencv, vllm
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V16.5.1
Mitigating Controls (NIST 800-53 r5) AI
si-11 directly requires error messages to avoid revealing exploitable information about the system or its data.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure SDLC practices directly prohibit writing sensitive data to logs; eliminating the weakness satisfies only one narrow slice of the control.
Log generation configuration can and should exclude sensitive data, but the control statement focuses on availability rather than content filtering.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Requiring de-identification and privacy controls before logs leave the organization reduces the chance that sensitive data inadvertently captured in logs becomes exposed to external parties.
Secure coding standards explicitly forbid exposing sensitive data in errors.
Security testing can detect error messages that leak sensitive information.
By defining what records must be kept, where, and for how long, the control discourages the inadvertent inclusion of sensitive information in logs or other externally accessible files that fall outside the formal record system.
Mandating deletion of temporary files and logs that may contain sensitive information prevents those artifacts from remaining accessible after the data is no longer needed.
When log entries are produced from masked data sets, the control prevents the inadvertent insertion of sensitive values into externally accessible log files.