Runway unveils Act-One, a revolutionizing character animation within Gen-3 Alpha platform

Runway has introduced Act-One, an innovative tool designed to generate expressive character performances within the Gen-3 Alpha platform. This advanced model utilizes video and voice inputs to craft compelling animations, marking a significant advancement in the application of generative models for both live action and animated content.

Act-One is capable of working with a wide range of reference images, maintaining realistic facial expressions while accurately translating performances into characters with differing proportions from the original source material. This adaptability facilitates new opportunities in creative character design and animation.

The model is particularly adept at producing cinematic and realistic outputs, demonstrating robustness across various camera angles while preserving high-fidelity facial animations. This capability enables the creation of multi-turn, expressive dialogue scenes, a complex task for previous generative video models. Users can now develop narrative content using just a consumer-grade camera and a single actor performing multiple characters from a script.

To accompany the release of Act-One, Runway has implemented a comprehensive suite of content moderation and safety features, ensuring responsible use of this powerful tool.