Gen-2 by Runway is a multi-modal AI system that has the ability to generate novel videos using various inputs such as text, images or video clips.

This tool can generate videos by applying the style and composition of an image or text prompt to the structure of a source video or even create videos using only textual prompts.

It can create high-quality and consistent videos as if they were filmed without actually filming anything at all.

Gen-2 offers eight modes that allow for different types of video generation.

Mode 1 is "Text to Video" which synthesis videos in any style using a text prompt, and Mode 2, "Text+Image to Video," generates videos using both an image and a text prompt.

Mode 3 generates videos using just a driving image, while Mode 4 transfers the style of any image or prompt to every frame of a video.

Mode 5, "Storyboard," turns mockups into fully stylized and animated renders, while Mode 6, "Mask," isolates subjects in a video and modifies them with simple text prompts.

Mode 7 is "Render," which turns untextured renders into realistic outputs using an input image or prompt, and Mode 8 is "Customization," which allows users to customize the model for even higher and more accurate results.

Gen-2 is preferred over existing methods for image-to-image and video-to-video translation, making it a valuable tool for various industries, especially those involved in video production and creative endeavors.