Moderation

Object containing moderation models designed to detect harmful content in text and images. These models are free to use and can identify various categories of potentially harmful content.

Properties

Link copied to clipboard

Omni-moderation is the most capable moderation model, accepting both text and images as input. It can identify potentially harmful content across multiple categories.

Link copied to clipboard

Text-moderation is a previous generation text-only moderation model. It can identify potentially harmful content in text across multiple categories.