Midjourney has introduced its inaugural video generation model, called 'V1,' marking the company’s expansion from static image generation to dynamic animation. The launch comes days after Disney and Universal filed a lawsuit against the startup for alleged copyright infringement.
V1 enables users to transform any image into short, 5-second video clips through either automatic animation or manual prompt inputs, where users can specify camera movements and actions to direct the clip’s flow.
Each V1 job produces four 5-second clips that can be extended to 20 seconds, with a pricing structure set at eight times the cost of an image job — which Midjourney claims is still 25 times cheaper than comparable services.
The system works with images generated within Midjourney or sourced externally, ensuring that the final videos maintain the distinctive aesthetic that has made the startup’s image models so popular among creators.
Founder David Holz described V1 as a foundational step towards Midjourney’s larger goal of real-time, open-world simulations that integrate image, video, and 3D modeling technologies.
While V1 does not currently include audio and focuses solely on image-to-video (I2V), it provides an intriguing look into how Midjourney might evolve towards more immersive creative tools in the future.
For full details and usage guidelines, visit the official update here.