Midjourney, a leading AI image generation startup, made waves with the launch of its highly anticipated AI video generation model, V1. This innovative image-to-video model enables users to transform images, generated by Midjourney’s other models or uploaded directly, into sets of four five-second videos. Exclusive to Discord and web-based initially, V1 is priced at $10 per month, making it the first of its kind widely accessible to all.
In the rapidly evolving landscape of AI video generation, Midjourney steps forward to compete with industry giants like OpenAI, Runway, Adobe, and Google. While many focus on commercial applications, Midjourney remains dedicated to nurturing creativity through its unique AI models. CEO David Holz envisions a future where AI models go beyond generating content for films and commercials, aiming for real-time open-world simulations.
The release of the V1 model coincided with legal action from Hollywood heavyweights Disney and Universal, who are concerned about potential copyright infringements through Midjourney’s AI-generated images. Despite these challenges, Midjourney continues to push boundaries, offering the V1 model at a premium cost to account for the higher resource consumption in video generation, especially in the “Relax” mode.
With customizable features, V1 empowers users to shape their video outputs to their liking. From automatic animations to manual text descriptions of specific movements, users have the flexibility to experiment. They can adjust camera and subject motions for varying visual effects, extending the default five-second videos up to 21 seconds by repeating sequences. Early feedback on V1 has been positive, praising its unique and somewhat surreal video outputs that set it apart from the competition.
