Midjourney joins the AI Video race with its first-ever video generation model
But the videos don't aim for realism like most offerings from the competition.
After years of focusing on AI-generated imagery, Midjourney is finally stepping into motion. This week, the company rolled out V1, its first-ever video generation model — and in doing so, officially joined the rapidly accelerating race to define the future of AI video.
V1 is an image-to-video tool that takes a single image and turns it into four short, five-second clips. Like Midjourney’s image models, it’s available only through Discord and limited to web use at launch.
It’s not breaking technical ground. OpenAI’s Sora, Runway’s Gen-4, Adobe Firefly, and Google’s Veo have been experimenting with similar tech for months. But where other companies are angling for commercial use cases like film production or advertising, Midjourney is targeting artists and designers.
The outputs reflect that focus. V1 doesn’t aim for realism. Its videos are more like animated daydreams, sometimes strange, sometimes stunning, but always unmistakably artificial.
In a blog post, founder David Holz described V1 as a stepping stone toward the company’s bigger vision: AI that can generate real-time, open-world simulations.
In the meantime, users can experiment with a set of controls that let them choose automatic or manual animation, adjust camera motion, and extend clips up to 21 seconds.
But that creativity isn't cheap. Video generations use up credits eight times faster than image generations, making Midjourney’s already-premium subscription plans even more limiting for casual users. Basic access starts at $10/month, while unlimited video use (in a slower queue) requires the $60 or $120/month plans. The company says it’s still evaluating pricing.
Kelechi Edeh
Boasting over 19 million users on its Discord server, it’s already one of the largest and most active communities in generative AI. That built-in base gives V1 a massive runway. If even a small percentage of users adopt video generation, it could supercharge revenue, expand its influence, and turn it into a serious contender in the video AI space.
However, V1’s debut also arrives under a legal cloud. Just days before the launch, Disney and Universal sued Midjourney over alleged copyright infringement tied to its image models. It’s the latest chapter in the industry’s legal standoff over who owns what when machines create art.
Still, the move into video feels inevitable. AI tools are becoming more immersive, more multimodal, and Midjourney, with its strong brand and massive user base on image creation, was never going to sit that out. V1 may not be the most advanced video model on the market, but it adds a new dimension to Midjourney’s creative toolkit. Whether it can scale that vision without losing its identity is the next question.
Kelechi Edeh
