One of the most popular generative art platforms, Midjourney, has introduced its first AI model for video creation. The new model, called V1, allows you to generate multiple video versions of up to 21 seconds long based on a single image.
One of the most popular generative art platforms, Midjourney, has introduced its first AI model for video creation. The new model, called V1, allows you to generate multiple video versions of up to 21 seconds long based on a single image.
As TechCrunch reports, the V1 model is already available to Discord users on the web. To use it, you need to upload an image or create one through Midjourney, after which the system will generate four five-second clips with different animations. The video can be extended up to 21 seconds, adding 4 seconds each time.
The model has two options: automatic — the system chooses how to animate the image, and manual — the user can describe the desired animation with text. Motion level settings are also available — from «low motion» to «high motion.»
Midjourney positions V1 not as just another video or advertising tool, but as a step towards fully-fledged real-time open simulation generation. The next stage will be the creation of 3D models and interactive environments.
Currently, V1 has limitations: the cost of generating video is eight times higher than that of images. The basic subscription costs $10 per month, but for unlimited video generation in the slow-motion «Relax» mode, you need a tariff of $60 or more. The company plans to review its pricing policy within a month.
The launch of V1 comes amid a high-profile lawsuit against Midjourney by Disney and Universal, which accuse the startup of infringing on copyrights, including those of characters from The Simpsons and Star Wars. It raises yet another question about the limits of legitimate use of AI in creative work.
V1 is Midjourney’s response to the success of video models from OpenAI (Sora), Runway (Gen-4), Adobe (Firefly), and Google (Veo 3). Unlike them, Midjourney remains true to its creative community on Discord, focusing on creativity rather than commercialization. The first demo videos have a distinctive «otherworldly» aesthetic that has made the brand famous in the world of artificial art.
As we previously reported, Midjourney has released the alpha version of its new AI model V7. It will have to compete with the image generator in ChatGPT, which went viral for its ability to create Ghibli-style photos.
