Runway has introduced a free alpha version of its Gen-3 artificial intelligence model for generating videos from promos and static images. However, it was not without limitations.

According to the announcement, the neural network excels at creating expressive human characters with a wide range of movements, gestures, and emotions. Gen-3 Alpha is trained to accurately identify key frames in a video and create smooth transitions.

You can use the neural network only through the Runway website. There you need to select the Turbo model, upload a start image, and then provide a detailed description of what you want to see.

The description itself should not exceed 500 characters. After that, Gen-3 Alpha generates a five- or ten-second video in high resolution. The generation time is 45 and 90 seconds, respectively.

There are no exact dates for the public release of Gen-3. The alpha version “will soon be available in the Runway product line with support for all existing modes (text-to-video, image-to-video, and video-to-video) and some new ones,” said Runway co-founder Germanidis.