62748 matches found
CVE-2026-12491 Vllm: vllm: image exif rotation & png trns transparency not normalized, causing mismatch between model input and expectations
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 informatio...
CVE-2026-12491
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 informatio...
CVE-2026-12491
CVE-2026-12491 affects the vLLM library used for LLM inference. The issue stems from improper handling of image metadata during image processing, specifically EXIF orientation and PNG transparency (tRNS). When converting images to RGB, transparency information may be discarded or remapped, causin...
CVE-2026-12491
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 informatio...
[SECURITY] Fedora 43 Update: openslide-4.0.0-14.fc43
The OpenSlide library allows programs to access virtual slide files regardless of the underlying image format...
CVE-2026-49218
A flaw was found in ImageMagick. A missing check in the DCM Digital Imaging and Communications in Medicine decoder allows a remote attacker to provide a specially crafted image with invalid dimensions. This can lead to crashes in other operations, resulting in a denial of service DoS for the...
CVE-2026-53461
A flaw was found in ImageMagick. An incorrect loop in the ICON decoder can lead to an out-of-bounds heap write. This vulnerability allows a remote attacker to cause a denial of service DoS by providing a specially crafted image file, leading to a system crash. Mitigation Red Hat is not aware of a...
CVE-2026-7850 WP Magnific Popup <= 1.0 - Author+ Stored XSS via href Attribute
The WP Magnific Popup WordPress plugin through 1.0 does not properly escape user-controlled link URLs before injecting them into the DOM when displaying image load error messages, allowing authenticated attackers with Author-level access or above to perform Stored Cross-Site Scripting attacks...
CVE-2026-7850
The WP Magnific Popup WordPress plugin (versions through 1.0 ) is vulnerable to Stored Cross-Site Scripting (XSS) . The root cause is the failure to properly escape user-controlled link URLs before injecting them into the DOM when the plugin displays image load error messages. This flaw allows au...
CVE-2026-7850 WP Magnific Popup <= 1.0 - Author+ Stored XSS via href Attribute
The WP Magnific Popup WordPress plugin through 1.0 does not properly escape user-controlled link URLs before injecting them into the DOM when displaying image load error messages, allowing authenticated attackers with Author-level access or above to perform Stored Cross-Site Scripting attacks...
ImageMagick < 6.9.13-50 / 7.x < 7.1.2-25 Multiple Vulnerabilities
The remote host has a version of ImageMagick installed that is prior to 6.9.13-50 or 7.x prior to 7.1.2-25. It is, therefore, affected by multiple vulnerabilities, including: - A missing validation in AcquireAlignedMemory could cause memory exhaustion conditions. CVE-2026-53460 - An incorrect loo...
ImageMagick 7.x < 7.1.2-24 Heap Buffer Over-write (CVE-2026-48724)
The remote host has a version of ImageMagick 7.x installed that is prior to 7.1.2-24. It is, therefore, affected by a heap buffer over-write vulnerability: - When using an image with mask the Floyd-Steinberg dithering method will cause a negative heap buffer over-write. CVE-2026-48724 Note that...
PT-2026-50243
The WP Magnific Popup WordPress plugin through 1.0 does not properly escape user-controlled link URLs before injecting them into the DOM when displaying image load error messages, allowing authenticated attackers with Author-level access or above to perform Stored Cross-Site Scripting attacks...
PT-2026-50278
Name of the Vulnerable Software and Affected Versions vLLM affected versions not specified Description An issue exists in the image processing logic where image metadata is improperly handled. Specifically, the software fails to call ImageOps.exif transpose to normalize EXIF orientation, causing...
PT-2026-50481
Name of the Vulnerable Software and Affected Versions Open-webui affected versions not specified Description An authenticated user can access files belonging to other users by exploiting a lack of ownership verification in the image processing path. When the POST endpoint "/api/chat/completions"...
PT-2026-50480
Name of the Vulnerable Software and Affected Versions Open WebUI versions prior to 0.9.6 Description An issue exists where the process picture url function in backend/open webui/utils/oauth.py performs URL validation only on the initial URL. Subsequently, it uses aiohttp.ClientSession.get without...
ROS-20260617-73-0072
The vulnerability of the console-based graphic editor ImageMagick is related to the execution of operations beyond the buffer boundaries in memory. Exploiting this vulnerability can allow an attacker to cause a system failure using a specially created TIFF file...
Google Pixel CPIF Driver Out-Of-Bounds Read / Write
The CPIF driver on Google Pixel has issues in linkreadgnssimage and linkloadgnssimage that lead to kernel out-of-bounds read and write from userland...
PT-2026-50610
Name of the Vulnerable Software and Affected Versions Drupal core affected versions not specified Description The JSON:API and REST modules allow image file uploads to image fields. The validation rules verify the file extension but fail to check the file MIME type Multipurpose Internet Mail...
CVE-2026-12491: Misinterpretation of Input
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 informatio...