Right after getting into legal trouble with entertainment giant Disney, AI startup has launched its image-to-video generation model V1. The new tool allows users to convert images into short video clips and is available to the users for free.
The new offering is available today via Midjourney's website, and it's available to all customers, including those who want to use a free account. Faster video rendering is limited to those users who pay for a subscription, which begins at $10 a month.
The move positions the platform in competition with AI-powered video creation. V1's launch is a significant step forward for Midjourney, which has spent the majority of its time creating visually rich static images. With the new tool, the start-up is going after users who want to animate the visuals they create, offering features that strike a balance between simplicity, customization, and affordability.
Despite the fact that the tool is available to all the Midjourney users free of cost at the preliminary level, one aspect of concern is the GPU usage by the tool, as the process of video generation is more resource-intensive, requiring eight times more GPU time per second of video. Hence, the users have to allocate GPU time carefully, especially when using the fast mode available under standard subscriptions.
The company has also provided Relax Mode to heavy users or creative professionals, but it is accessible only to Pro-tier users and higher. This mode allows unlimited videos at no extra cost, but with longer waiting times, it can at times be up to 10 minutes for one animation. "The pricing model still gives you a massive cost advantage," the company says—reportedly being over 25 times cheaper than any of the current AI video generation tools on the market.
Though the new tool is not far from controversy, like many other artificial intelligence models. Midjourney was sued for copyright infringement by major media companies, including Disney and NBCUniversal. The lawsuit claims that the platform enables unauthorized derivative works by reproducing characters and images from popular franchises.
The studios said Midjourney has essentially become a "virtual vending machine" for pirated offerings, citing instances using characters from Star Wars, Toy Story, Marvel, and The Simpsons. The company has not publicly addressed the lawsuit or issued a public statement about the legal questions raised about V1.
Midjourney founder David Holz has called the video model a "stepping stone" to more sophisticated applications, such as real-time, three-dimensional environments that users can interact with.
As far as the actual animation process is concerned, Midjourney V1 is designed to be user-friendly as seen below:
- Users begin by uploading an image—either created within Midjourney or imported from another source.
- After the image is generated, an "Animate" button lets users transform the static image into a video.
- The system produces four 5-second video clips from the chosen image.
- These clips can be extended by four-second increments, up to a total of 21 seconds.
- Currently, the videos are silent, with no audio features supported.
- At present two animation modes are available: First, automatic mode, which applies system-suggested motion effects to the image. Second, manual mode lets users guide the animation through descriptive prompts.
- Users can also choose between two camera movement styles: Low Motion keeps the camera largely stationary, and High Motion introduces dynamic movement of both the subject and the camera.
This new offering introduces an intuitive way to blend static visuals with motion, enhancing creative storytelling possibilities on the platform.
Netizens were quick to upload the videos generated by the tool, which shows the kind of interest this tool has generated as seen above.
