Overview
- V1 converts still images into four separate five-second videos that users can extend up to 21 seconds.
- The model is available on the web and via Discord with subscriptions starting at $10 per month, and video rendering jobs consume eight times more GPU time than image generations.
- Users can select automatic or manual animation modes and toggle between low and high motion settings to control camera and subject movement.
- Disney and Universal sued Midjourney days earlier, alleging its AI was trained on their copyrighted characters without permission.
- According to CEO David Holz, V1 serves as a stepping stone toward developing real-time open-world simulation AI and subsequent 3D rendering models.