Midjourney announced the launch of the first version of its V1 model for video generation for all users. The service now features an “Image-to-Video” function, allowing previously created images to be animated using the “Animate” button. Users can choose an “automatic” mode, where AI independently creates movement, or a “manual” mode, where they can describe the desired animation and scene development in detail.
The new interface offers “high motion” and “low motion” settings. “High motion” is suitable for dynamic clips where both the camera and objects move, while “low motion” is for calm scenes with minimal changes. Each video clip lasts about 5 seconds and can be extended four more times, resulting in up to 20 seconds of animation in total.
In addition to images created in Midjourney, users can upload their own pictures, select them as a “starting frame,” and add a description of the movement. This provides the opportunity to use archival photos or personal illustrations.
The feature is available only in the web version, and the cost of one video request is approximately eight times higher than creating an image. However, with one request, the user receives four videos, each 5 seconds long, making the price per second of animation comparable to image enhancement. A special “relax” mode for video is being tested for “Pro” subscribers.
