CVE-2026-12491
MEDIUMDescription
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.
Is your site exposed to CVE-2026-12491?
Run a free security scan — no signup, results in seconds.
CVSS v3.1 Score
EPSS — Exploit Prediction
EPSS estimates the probability that this vulnerability will be exploited in the wild within the next 30 days. A higher score means more likely to be exploited.
Weakness Type (CWE)
References
Frequently Asked Questions
What is CVE-2026-12491? +
How severe is CVE-2026-12491? +
How do I check if I'm vulnerable to CVE-2026-12491? +
Related Vulnerabilities
pgAdmin 4's Import/Export Data tool builds a psql \copy (...) command line by interpolating a user-supplied SQL query into a …
The fix for CVE-2026-12045 in pgAdmin 4 9.16 required the LLM-supplied query passed to the AI Assistant's execute_sql_query tool to …
WOLFBOX Level 2 EV Charger MCU Command Parsing Misinterpretation of Input Remote Code Execution Vulnerability. This vulnerability allows network-adjacent attackers …
An unhandled exception in danny-avila/librechat version 3c94ff2 can lead to a server crash. The issue occurs when the fs module …
A flaw was found in libsoup. The HTTP/2 server in libsoup may not fully validate the values of pseudo-headers :scheme, …
A Cross-Protocol Scripting vulnerability is found in Apache Kvrocks. Since Kvrocks didn't detect if "Host:" or "POST" appears in RESP …