Midjourney, known for its impressive image generation tools, has just launched V1 of its Video Model, marking its first venture into video creation. This release, dubbed “Image-to-Video,” lets users transform static images into short, dynamic video clips. Available at $10 per month, now available on the web, offering an accessible way for users to experiment with AI-driven video.

The process is straightforward: users start with an image created in Midjourney or upload one from elsewhere. From there, they can hit the “Animate” button to generate a video. Midjourney offers two animation modes—automatic and manual. The automatic mode includes “high motion” for lively scenes with moving subjects and cameras, and “low motion” for calmer, more static environments. The manual option lets users describe specific movements and scene developments. Each animation job produces four 5-second videos, with the option to extend clips up to four times, roughly 4 seconds per extension.

Midjourney emphasizes affordability, noting that a video job costs about eight times more than an image job, comparable to the cost of an image upscale, or roughly one image’s worth per second of video. This makes it significantly cheaper than other video generation tools currently available. A “video relax mode” is also being tested for Pro subscribers and above, though details on its functionality are still emerging.

The company is candid about the challenges of scaling this technology, particularly around server capacity and operational costs. They plan to monitor usage over the next month to ensure sustainability, which may lead to adjustments in pricing or access. For now, the focus is on delivering a tool that’s fun, user-friendly, and visually appealing, with the added flexibility of animating external images as starting frames.

Midjourney’s broader vision is ambitious: they aim to build toward real-time, open-world simulations where users can navigate 3D spaces and interact with dynamic environments and characters. This video model is a key stepping stone, following their work on image models and paving the way for future 3D and real-time systems. The company also hints at upcoming improvements to their image models, informed by insights gained from developing this video tool.

For creators, hobbyists, or anyone curious about AI-driven video, Midjourney’s V1 Video Model offers a practical and engaging way to bring images to life. Users are encouraged to explore responsibly and share feedback as the platform evolves.