In a notable move, Midjourney has launched the first variant of its video generation model to the public, ensuring a smooth and intuitive user experience.
Currently, the recently launched tool is able to generate short videos based on images uploaded on the app; however, the company plans to introduce a range of unique capabilities soon.
In a video posted on X (formerly Twitter), Midjourney founder David Holz called this first version only “a stepping stone” as the company gears up to create “models capable of real-time open-world simulations.”
Once you have generated an image using Midjourney, tap the latest “animate” button to create a 5-second clip based on a text prompt.
It will also offer an option to add an image you uploaded to the service as a “starting frame” for a video.
Moreover, it allows users to extend an animation by four seconds up to four times, generating for a 21-second-long video in total.
Midjourney's V1 availability and pricing
The company further stated that it will cost “about 8x more for a video job than an image job,” adding up to around 'one image worth of cost’ per second of video."
Midjourney’s AI video generator is currently accessible on the web and via the startup’s Discord server.
To access it, users are required to subscribe to it which starts at $10 / month for 3.3 hours of “fast” GPU time.
