← All CVEs

CVE-2026-12491

MEDIUM 4.8

Published 2026-06-17 · Last modified 2026-06-17

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.

NO EXPLOITATION SIGNALS

No known exploitation, public exploit, or elevated probability at this time. Track for changes.

Exploitation likelihood

0.2%chance of exploitation in 30 days · 15th percentile

○ In CISA KEV ○ Public exploit / PoC

Impact if exploited

4.8CVSS 3.1 · MEDIUM

  • ConfidentialityNone
  • IntegrityLow
  • AvailabilityLow

What an attacker needs

  • Access: Reachable over the network — no local access needed
  • Privileges: No account or privileges required
  • User interaction: No user interaction needed
  • Complexity: Needs a race window or specific setup

✓ lowers the bar for an attacker · ⚠ raises it

Affected

Vendors Red Hat

Products Red Hat Ai Inference Server Red Hat Enterprise Linux Ai (Rhel Ai) 3 Red Hat Openshift Ai (Rhoai)

Weakness (CWE)

  • CWE-115

CVSS vector

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

References

Technical & other

Sources: NVD · CVE.org · EPSS