Runway unveils Gen-4, its new model for advanced, consistent and dynamic video generation

Runway has introduced Gen-4, its newest series of AI models for media and video generation, available to those with paid and enterprise subscriptions. This new Image-to-Video model outperforms the previous Gen-3 Alpha by offering better fidelity, dynamic motion, and improved adherence to prompts.

Gen-4 is particularly adept at creating dynamic videos with realistic motion, ensuring consistency in characters, objects, and visual style across various shots. Users can input a single reference image along with descriptive instructions to generate consistent visuals from different perspectives, enhancing narrative continuity and creative control.

The Gen-4 model tackles common challenges in AI-video generation, especially in maintaining visual and stylistic consistency, and includes an advanced understanding of the world for better prompt adherence.