Runway launches Frames image generation model
Runway has introduced Frames, a new base model for image generation built around stylistic consistency, and is rolling it out gradually to users of Gen-3 Alpha and its API.
Runway announced Frames, which it calls its newest base model for image generation. The company says the model is "a big step forward in stylistic control and visual fidelity."
The main pitch is consistency. According to Runway, Frames can set a specific look for a project and reliably produce variations that keep that aesthetic, while still leaving room for broader creative exploration. The company describes this as letting users "architect worlds" with a defined point of view, look, and atmosphere.
Access is arriving gradually inside Gen-3 Alpha, Runway's video generation platform, and through the Runway API. Runway did not give a timeline for wider availability.
The announcement has no benchmarks, technical details about architecture or training, or pricing. Instead, Runway shows sample "worlds" in different styles. These include cinematic portraiture, 1980s practical-effects creature makeup, 1970s-style retro anime, 35mm flash photography, collage built from vintage clippings, landscape photography, and still lifes of everyday objects.
Runway says Frames ships with "multiple content moderation and safety procedures" and is covered by its Foundations for Safe Generative Media framework. The announcement does not say what those measures are.
Without published evaluations, it is not yet possible to compare Frames' fidelity or style control with other image models.