MidJourney’s new video AI animates generated images with impressive detail and smooth camera movements, making it suitable for engaging online content despite some motion quirks and lower resolution. Operating within MidJourney’s subscription plans, it offers customizable animation options, efficient previews, and an extend video feature, positioning it as an affordable and promising tool for creative video generation.
The video provides an in-depth look at MidJourney’s new video AI model, highlighting its surprising quality and ease of use. Unlike text-to-video models, MidJourney’s video AI starts with an image generated by the platform and animates it, maintaining impressive detail consistency throughout the motion. The presenter showcases various examples from MidJourney’s website, noting the model’s strengths in creating smooth camera movements around characters and landscapes, though some animations exhibit quirky or unnatural motions, such as erratic background elements or inconsistent object behavior. Despite these imperfections, the model’s output is deemed suitable for online content like memes, where attention-grabbing visuals are key.
Technically, the video AI operates under MidJourney’s existing pricing structure, with video generation consuming GPU time from users’ subscription plans. The standard plan offers limited video generation hours, while the pro plan provides unlimited video and image creation for a higher monthly fee. Users can choose between low and high motion modes, with high motion requiring significantly more GPU time but offering more dynamic animations. The platform also supports both automatic and manual prompt modes, allowing users to customize video content by specifying motions like zooms or rotations, and even upload their own images as starting points for animation.
The video resolution is relatively low compared to traditional video standards, but MidJourney compensates by offering higher-resolution downloads optimized for social media. Users can preview videos efficiently by scrubbing through frames with keyboard and mouse controls, and advanced options like “raw” mode reduce AI creativity for more prompt-accurate results. Additionally, an extend video feature lets users continue animations either automatically or by manually inputting prompts to guide the next sequence, providing flexibility in storytelling or content creation.
The presenter also shares personal experiments with the video AI, demonstrating its ability to animate complex scenes such as dancing characters and dynamic backgrounds. While the model struggles with synchronizing head and body movements—common in AI-generated dance animations—it still produces visually compelling clips. Comparisons with other AI video models reveal that MidJourney’s offering holds up well, particularly in maintaining cinematic qualities like camera motion and environmental effects, despite occasional motion inconsistencies.
In conclusion, MidJourney’s video AI is praised for its impressive capabilities, affordability, and potential for growth as the technology evolves. The presenter encourages viewers to consider the value of unlimited video generation under the pro plan and to experiment with the tool’s creative possibilities. The video ends with an invitation for viewers to share their thoughts on the model and pricing, signaling ongoing interest in the development of AI-driven video creation tools.