Purpose: This tool is used to quickly determine whether an image is suitable for viewing in the workplace (Safe For Work). It adapts Stable Diffusion's built-in safety checker into a filter that can process arbitrary images, not just AI-generated images.
Main capabilities: Supports uploading PNG or JPG images, up to 50MB each; after upload, it returns a determination of whether the image is safe. It is officially described as an extremely fast machine learning NSFW image filter, based on an open-source model.
How to use: After opening the webpage, click upload or drag and drop an image into the designated area; no account creation is required.
Use cases: For situations where images need to be quickly screened for explicit content, such as personal self-checks or simple content filtering demonstrations.
Limitations: The official description does not state the accuracy of the determination, supported languages, whether images are saved, or API or batch processing capabilities; only PNG and JPG are supported, and files must not exceed 50MB. Developer Miguel Pascual presents it as a public demo project.
