Midjourney launches a video generator AI

Midjourney released the First version of his video generation model to the public. For the moment, the tool can generate short videos depending on the images downloaded or created on the platform, but Midjourney plans to deploy more capacities in the future.

After creating an image using Midjourney, the service will display a new “animated” button that users can press to create a 5 -second clip based on a text prompt. It will also display the option of adding an image that you have downloaded on the platform as a “starting frame” for a video. The tool will generate a generic prompt which “simply makes things” move “, but a” manual “button allows users to describe what they want that the movement looks like.

Users can extend an animation From four seconds to four times, making a video of 21 seconds in total. There are also high and low movement parameters that control if the subject and the camera move, or simply the subject.

Midjourney’s AI video generator is currently only available on the web and via the startup discord server. It requires a subscription to the service, which starts at $ 10 / month for 3.3 hours of “fast” GPU (around 200 generations of images). The startup indicates that it will invoice “about 8 times more for video work than an image work”, adding to about “a cost image” per second of video “.

Midjourney is currently the subject of a Disney and Universal trialWho quoted the prospect of launching a video generator as a special concern. He maintains that Midjourney offers a “virtual automatic distributor, generating unauthorized endless copies of the work protected by the copyright of Disney and Universal”. The current video generation model was Announced for the first time in JanuaryAnd Disney and Universal argued that its training process meant that “Midjourney is most likely already violates the works protected by the complainants”.

In an article announcing the generator, the founder of Midjourney, David Holz, says that this first version is just “a springboard” while the startup works on the creation of “models capable of simulations in time open in real time”. Google,, OPENAIAnd Meta also launched AI video generators, which can all generate videos with text prompts.