Midjourney Unveils V1: Its First AI Video Generation Model
AI-summarised brief · reviewed before publication
On June 18, 2025, Midjourney debuted its highly anticipated AI video generation model, V1, marking its entry into the competitive AI video creation space. The company also shared a roadmap outlining future enhancements for the model, signaling ambitious plans to expand its capabilities. The V1 model, priced at $10 per month, is designed to be accessible, user-friendly, and visually striking, making it what Midjourney calls “the first video model for everyone.” The announcement, shared via a post on X, emphasized its ease of use and creative potential. Unlike competitors such as OpenAI’s Sora, Runway’s Gen 4, Adobe’s Firefly, and Google’s Veo 3, which often prioritize commercial applications, Midjourney’s V1 focuses on empowering creative users. CEO David Holz highlighted the company’s long-term vision, stating in a blog post that V1 is a step toward building AI models capable of “real-time open-world simulations,” far beyond generating standard video content like B-roll for films or advertisements. V1 is an image-to-video model, allowing users to upload an image—either their own or one generated by Midjourney’s other models—and animate it into a five-second video. Users can customize outputs with settings like automatic animation for random motion or manual text prompts to define specific movements. Options for “low motion” or “high motion” further adjust camera and subject dynamics. While videos are initially limited to five seconds, users can extend them by four seconds up to four times, reaching a maximum of 21 seconds. Early demos showcase a distinctive, dreamlike aesthetic, consistent with Midjourney’s reputation for producing otherworldly visuals rather than hyperrealistic ones. The launch comes amid legal challenges for Midjourney. Just a week prior, Disney and Universal filed a lawsuit alleging that Midjourney’s AI image models generated content featuring their copyrighted characters, such as Homer Simpson and Darth Vader. The suit reflects broader industry concerns about AI tools potentially devaluing creative work or relying on copyrighted material for training. Despite Midjourney’s efforts to position itself as a creativity-driven platform, these accusations highlight ongoing tensions in the AI content generation space. Currently, V1 is accessible exclusively through Midjourney’s website, with video generation costing approximately eight times more than image generation. A “video relax mode” is available for Pro+ subscribers, and the model supports external image uploads. Posts on X indicate strong enthusiasm for V1’s cinematic and dreamlike outputs, with generation times ranging from three to five seconds for videos at 1024x576 resolution. Midjourney’s V1 launch positions it as a formidable player in AI-driven video creation, blending accessibility with a focus on artistic expression, though its legal battles and the high cost of video generation may shape its path forward.
