Midjourney releases its initial AI video creation model, V1

On June 18, 2025,Midjourney V1, a leading name in AI image generation, introduced its first AI video generation model, Midjourney V1. This release marks a significant step for the San Francisco-based startup, expanding its creative tools from static images to dynamic video content.

Features and Functionality

The V1 model, accessible via Midjourney’s web interface and Discord, transforms a single image either generated by Midjourney V1 or uploaded by the user and into up to four five-second video clips. Users can extend these clips by four seconds up to four times, creating videos as long as 21 seconds. The model offers two modes:

• Automatic Mode: The AI determines the animation style and movement.
• Manual Mode: Users input text prompts to customize the animation, with options for “low motion” (ideal for ambient scenes) or “high motion” (suitable for dynamic sequences).

Additional settings allow control over camera and subject movement, enabling tailored outputs. However, V1 videos are currently limited to 480p resolution at 24 frames per second, lack audio, and do not support upscaling. Despite these constraints, testers have praised the model’s coherence, inherited from Midjourney’s V6.1 image model, though high-motion scenes may exhibit flickering.

Pricing and Accessibility

V1 is available to all Midjourney subscribers, starting at the $10-per-month Basic plan. Video generation costs eight times more than image generation, consuming credits faster. For example, a single video job produces four five-second clips but uses significantly more processing resources. A “Video Relax Mode” is being tested for Pro-tier subscribers and above, allowing video creation without depleting fast-processing minutes, potentially reducing costs.

Midjourney’s V1pricing undercuts competitors like OpenAI’s Sora and Google’s Veo 3, making V1 an affordable entry point for creators. The company emphasizes accessibility, noting that its 20 million subscribers can immediately use the “Animate” button to bring images to life.

Strategic Vision

Midjourney’s CEO, David Holz, described V1 as a “stepping stone” toward creating AI models capable of real-time, open-world simulations. The company aims to develop 3D rendering and interactive simulation tools, building on lessons from V1. Holz emphasized that V1 is not just for generating short clips for ads or films but part of a broader ambition to redefine visual storytelling.

Competitive Landscape

V1 enters a competitive AI video generation market, trailing behind models like Meta’s Luma Dream Machine, and Google’s Gemini in scope. Unlike Google’s Veo 3, which supports text-to-video and audio generation, V1 focuses solely on image-to-video (I2V) workflows. However, its integration with Midjourney’s robust image engine ensures consistent, stylistically rich outputs, appealing to its existing user base.

Challenges and Controversies

The launch comes amid legal scrutiny. On June 11, 2025, Disney and Universal Studios filed a lawsuit against Midjourney, alleging copyright infringement in its image generation models. The studios claim Midjourney’s AI was trained on their intellectual property, producing unauthorized copies of characters like Yoda and the Minions. This lawsuit, filed just a week before V1’s release, raises questions about the training data for Midjourney’s video model and potential future restrictions.

Community Reception

Early feedback on X highlights excitement among creators. Users praised V1’s ease of use and visual quality, with some calling it “insanely good” for a first iteration. Others noted its limitations, such as low resolution and lack of audio, but expressed optimism about future updates, anticipating improvements by the time the model reaches V7, like Midjourney’s image generator.

Looking Ahead

Midjourney’s V1 is a bold entry into AI video generation, leveraging the company’s strengths in visual aesthetics while navigating a crowded market and legal challenges. As Midjourney refines its technology and scales its infrastructure, V1 could reshape how creators animate and share stories, paving the way for more immersive AI-driven experiences.