Midjourney Launches V1, Its First AI-Powered Image-to-Video Generator

86

New tool enables customisable, animated video clips from still images

San Francisco-based AI lab Midjourney has announced the release of V1, its first-ever text-to-video generation model, marking a major step forward in AI-driven creative tools. Unveiled on June 18, V1 allows users to transform still images—either uploaded manually or generated through Midjourney—into five-second animated video clips using artificial intelligence.

Each image processed through V1 results in four unique five-second clips, and users have the option to extend each animation up to 20 seconds. While the company has not confirmed whether the clips will include sound, the broader aim is to eventually support real-time, open-world simulation capabilities. According to Midjourney’s leadership, the long-term vision is to develop systems that generate live imagery dynamically, responding to real-time inputs.

Dual Animation Modes and Camera Customisation

V1 offers users two animation pathways. In Automatic mode, the AI suggests a motion prompt to bring the image to life, while Manual mode allows users to define the movement and scene development through written prompts. Additionally, users can select from two camera settings: Low Motion, which simulates a stationary camera with subtle movement, and High Motion, which incorporates active motion in both the camera and subject.

One of the key highlights of V1 is its accessibility. The tool is available across all user tiers, including free accounts. However, video generation consumes significantly more computational resources than still images—eight times more GPU time, according to the company. Despite this, Midjourney claims V1 is substantially more cost-effective than current alternatives in the AI video market, estimating it to be over 25 times cheaper.

Users can access V1 through two operational modes: Fast Mode, which utilises monthly GPU time allocations, and Relax Mode, currently available to Pro-tier subscribers. While Relax Mode offers unlimited generation capacity, it comes with longer processing times of up to 10 minutes due to queued tasks.

Midjourney’s V1 rollout positions the company at the forefront of the evolving AI video landscape, offering a blend of affordability, creativity, and control to users across skill levels.