Midjourney , a company known for generating images using artificial intelligence, has announced the launch of its first video-making model, the V1 .
It’s an image-to-video model that allows users to upload an image (or generate one through other Midjourney models) and get four 5-second videos . Like the company’s previous products, V1 currently works exclusively through Discord and is only available on the web.
Midjourney is effectively competing with giants like OpenAI's Sora, Adobe's Firefly, and Google's Veo. At the same time, the company emphasizes that its main goal is not just to create videos for advertising or film, but to develop tools for realistic simulations in an open world in real time .
The launch of V1 comes amid lawsuits from Disney and Universal, which accused the company of using its generated images to infringe on copyrighted characters. Hollywood has increasingly criticized AI services, fearing they could impact the creative professions market.
In terms of cost, video generation will be more expensive than image generation. The minimum subscription to access V1 is $10 per month, but unlimited use requires a Pro ($60) or Mega ($120) subscription.
The model has a number of settings: users can choose automatic animation or manually describe what should happen in the frame. There are also high or low motion modes available, and the video can be extended to 21 seconds.
Early demos of the V1 show off Midjourney's signature surreal aesthetic. User feedback has been mostly positive, though it's still hard to gauge how well the V1 will compete with video models that have been on the market for months or years.