Predictive Vulnerability Intelligence.

Product

  • Product
  • Pricing
  • Documentation

Company

  • About
  • Partnerships
  • Blog
  • Support

Legal

  • Terms
  • Privacy
  • Data Licensing

© 2026 FAUCET Technologies LLC. All rights reserved.

CVE-2026-12491

22
FAUCET Score

A flaw was found in vLLM, an open-source library for large language model inference. This vulnerability arises from improper handling of image metadata, specifically EXIF orientation and PNG transparency (tRNS) data, during image processing. When images are converted to RGB, transparency information may be implicitly discarded or remapped, leading to unexpected rendering of transparent pixels and distortion of input content. This can result in the model misinterpreting image content, potentially affecting the integrity of processed data.

First published: Jun 17, 2026Last modified: Jul 7, 2026

Impacted Technologies

VendorProductVersion(s)CPE
Red HatRed Hat AI Inference Server
All Versions ImpactedCNA affecteddefault affected
Red HatRed Hat Enterprise Linux AI (RHEL AI) 3
All Versions ImpactedCNA affecteddefault affected
Red HatRed Hat OpenShift AI (RHOAI)
All Versions ImpactedCNA affecteddefault affected
Vllm-ProjectVLLM
>= 0.11.0, < 0.24.0CNA affecteddefault unaffected

CVSS Data

CVSS version used by this source: 3.1

4.8MEDIUM

CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L

Attack Vector
NETWORK
Attack Complexity
HIGH
Privileges Required
NONE
User Interaction
NONE
Scope
UNCHANGED
Confidentiality Impact
NONE
Integrity Impact
LOW
Availability Impact
LOW
Exploitability Score
2.2
Impact Score
2.5
CvssVersion
3.1

Exploit Intelligence

EPSS Score
0.24%
Probability of exploitation in next 30 days
EPSS Percentile
15.1%
Percentile rank of EPSS score among Peer Group
As of 2026-07-27
Model: v2026.06.15
This CVE's current EPSS score of 0.0024 is in the 2nd percentile among its peer group of 19,956 CVEs.

Social Chatter

No social media mentions found for this CVE.

The average CVE in this peer group has 0.0 Twitter, 0.0 Reddit, 0.0 Bluesky, 0.0 Mastodon, and 0.2 GitHub mentions.

Media Mentions

No media coverage found for this CVE.

The average CVE in this peer group has 0.0 InfoSec Media, 0.0 Vendor Blog, and 0.0 Security Researcher mentions.

Remediation

Patch Available

Vendor Patches (1)

pippatch availablevia ghsa
Product: vllmFixed in: 0.24.0

Vendor Advisories (1)

pipGHSA-8jr5-v98p-w75mmedium

vLLM: image EXIF Rotation & PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and Expectations

Jun 17, 2026

References

access.redhat.com / security/cve/CVE-2026-12491
bugzilla.redhat.com / show_bug.cgi