A first frame to video API allows developers to define how an AI-generated video begins by providing an initial visual reference. Instead of generating every frame from a text prompt alone, developers can use a specific image as the starting point and let the video model generate motion and subsequent scenes based on that reference. This workflow can be useful for maintaining a specific composition, character appearance, visual style, or product presentation while introducing natural movement. It can support a variety of applications, including animation, storytelling, product demonstrations, social media content, advertising, and other creative video projects. By giving developers more control over the opening frame, a first frame to video API can also make it easier to build consistent and repeatable video generation workflows within websites, applications, and creative platforms.