Gen-1

This model, released by Runway Research in February 2023, generates new videos from existing footage. It applies the composition and style of an image or text prompt to the structure of a source video, so a user can transform clips without filming anything new. The model works in five modes. Stylization transfers the style of an image or prompt to every frame. Storyboard turns mockups into stylized, animated renders. Mask applies changes to selected regions of a video and leaves the rest untouched. Render turns untextured 3D renders into realistic output using an input image or prompt. Customization fine tunes the model for higher fidelity results on a specific subject. Runway reports that in its user studies, results were preferred over Stable Diffusion 1.5 by 73.53 percent of participants and over Text2Live by 88.24 percent. The page links to the accompanying paper by Patrick Esser, Johnathan Chiu, Parmida Atighehchian, Jonathan Granskog and Anastasis Germanidis. It suits filmmakers, designers and video creators who want to restyle or re render existing footage. The page invites readers to try it in Runway, and Runway also offers iPhone and Android apps. The research page lists no price specific to this model. Runway's general pricing page shows a Free plan at $0 per month with 125 one time credits, and paid individual plans of Standard at $12 per month, Pro at $28 per month and Max at $76 per month when billed yearly, against $15, $35 and $95 billed monthly. Team, Enterprise and API options are also listed. That pricing page names Runway's current models rather than this one.