AI startup Runway, which created the renowned Gen-3 Alpha picture generator, has released a new foundational model that “excels at maintaining stylistic consistency while allowing for broad creative exploration,” according to the company.
The new format, called Frames, allows users to create photos on a wide range of subjects while maintaining a consistent visual style and ethos.
Frames adhere to the artistic style dictated by the user, whether it’s imitating ’80s comedy films like Flash Gordon and Xanadu, aping the format of ’90s-era 35mm disposable cameras or retro anime, creating sweeping landscapes or precisely framed still-life photos.
“With Frames, you can begin to architect worlds that represent very specific points of view and aesthetic characteristics,” the business stated in its announcement post on Monday. “The model allows you to design with precision the look, feel, and atmosphere of the world you want to create.”
Frames will not replace but rather complement, the current Gen-3 Alpha generation model. “We’re gradually rolling out access inside Gen-3 Alpha to allow you to build more of your worlds within a larger, more seamless creative flow,” according to the business.
Gen-3 Alpha is a relatively new model, having debuted in June 2024. It is designed for large-scale multimodal training and represents a “major improvement in fidelity, consistency, and motion over Gen-2, as well as a step towards building General World Models,” according to the company. The model was recently modified to provide more precise camera controls as well as video-to-video production capabilities.
The Gen-3 alpha has been trained on thousands of YouTube videos, prompting claims of copyright violation from YouTube content creators. 404 Media ran tests and discovered that mentioning a certain creator (say, MrBeast) in the prompt caused the system to generate graphics in that author’s style.