Midjourney has officially entered the AI video arena with a community-wide rollout of its new Image-to-Video tool, offering a scalable, affordable way to animate still images and paving the way toward fully interactive, real-time simulations. (Source: Image by RR)

Video Creation Brings a New Layer of Depth to Midjourney’s Visual Ecosystem

In a bold move signaling the next phase of AI media evolution, Midjourney has launched Version 1 of its much-anticipated Video Model. The release marks a strategic pivot from static image generation toward dynamic, animated content, moving closer to the industry’s long-term goal: real-time, open-world AI simulations. This vision entails intelligent systems capable of rendering immersive 3D environments in real time—enabling users to interact with both scenery and characters as they unfold fluidly on screen. While still in its early stages, the infrastructure Midjourney is building could reshape how we engage with visual media.

The Video Model debut, made available to the entire Midjourney community, introduces an “Image-to-Video” workflow that leverages the platform’s existing strengths in image generation. Users can now animate images with a single click, choosing between an “automatic” mode—where motion prompts are generated for them—and a “manual” option that lets them direct how scenes move and evolve. This dual-mode flexibility, as noted in midjourney.com, allows for both ease of use and fine-tuned creative control, ensuring broad appeal across experience levels.

Further enhancing usability, the platform offers both “high motion” and “low motion” settings, tailoring animations for either kinetic action or subtle, ambient movement. Each clip begins at five seconds and can be extended in four-second increments, up to 20 seconds total. Notably, the tool isn’t limited to images generated on Midjourney—users can upload external images, mark them as starting frames, and apply motion prompts to breathe life into static visuals. This positions Midjourney as not just an art generator, but a growing hub for animated digital storytelling.

For now, the rollout is web-only, and pricing has been set at approximately eight times the cost of a still image generation—an impressive feat, considering the company claims this to be over 25 times cheaper than previous video AI offerings on the market. Each job yields four 5-second clips, making the cost roughly equivalent to a single upscale per second of video. With a relaxed mode in testing for Pro subscribers, the company is clearly experimenting with scalability. This launch could represent a turning point in consumer AI creativity, merging accessibility with high-tech ambition in a way few others have achieved.

read more at midjourney.com