Midjourney officially launched its V1 Video Model in June 2025, marking the company’s first foray into AI video generation. The launch was announced with characteristic simplicity on social media, describing the tool as “enjoyable, straightforward, and visually appealing”.
I have a paid subscription to Midjourney, and in making the announcement, their CEO, David Holz, posted a message in the user dashboard. Here’s what it said (in summary).
Summary of the CEO’s Message
- Midjourney’s long-term goal is to create real-time, interactive AI systems that generate 3D worlds and simulations you can move through and interact with.
- To achieve this, they’re developing image, video, 3D, and real-time models, releasing them individually before combining them into a unified system.
- They’re releasing Version 1 of their Video Model, available to all users. This is a major step toward their larger vision.
How It Works
Unlike many competitors that offer text-to-video generation, Midjourney’s approach is distinctly image-first. The system works in two primary ways:
Image-to-Video Process
- Users can animate any image previously generated within Midjourney by clicking the “Animate” button
- External images can also be uploaded and used as starting points. The system generates four different 5-second video variations from each source image
- Videos can be extended in 4-second increments up to four times, creating clips up to 21 seconds total
Motion Controls
The platform offers two motion settings:
- Low motion: Suitable for ambient scenes where only the subject moves
- High motion: Both the subject and the camera are in motion
- Users can choose between automatic animation or manual control with custom motion prompts
Pricing and Accessibility
Midjourney’s pricing strategy appears deliberately competitive. At $10 per month for the basic plan, it significantly undercuts major competitors:
- OpenAI’s Sora: $20-$200/month
- Google’s Veo: $249/month
- Adobe Firefly: $29/month
- Runway Gen4: $12/month
However, video generation is resource-intensive, costing 8 times more GPU time than image generation – essentially “the cost of one image per second of video”.
Technical Specifications and Limitations
The current V1 model has several constraints that distinguish it from competitors:
Current Capabilities
- Resolution: 480p (with developers working on an upscaler)
- Duration: 5-21 seconds maximum
- Platform: Web-only at launch, with Discord integration
- Format: Optimized for social media platforms like Instagram, TikTok, and Twitter
Notable Limitations
- No text-to-video capability – requires an image starting point
- No audio generation – videos are silent
- Limited prompt control compared to competitors focused on commercial filmmaking
Strategic Positioning
Midjourney’s approach reflects a deliberate strategic choice. While competitors like OpenAI’s Sora and Google’s Veo target commercial filmmakers and studios with highly controllable tools, Midjourney remains focused on creative accessibility.
CEO David Holz described V1 as merely “a stepping stone” toward the company’s ultimate goal of creating “AI models capable of real-time open-world simulations”. The roadmap includes plans for 3D rendering capabilities and real-time AI models.
About the Podcast Host
Joanne Sweeney is a specialist in public-sector marketing and digital communications. She’s the author of Public Sector Marketing Pro and the creator of the Professional Diploma in AI for Public Sector Marketing. With years of experience in training, consulting, and advocating for digital transformation, Joanne provides insight and guidance for government agencies navigating the digital landscape. She is also the host of the award-winning Public Sector Marketing Show podcast.
