Midjourney rolls out V1 AI video generation model
Midjourney unveiled its V1 AI video generation model on Wednesday. The image-to-video model, accessible through Discord, produces four five-second videos from a user-uploaded or Midjourney-generated image and will cost 8x more for a video generation than a typical image generation.
CEO David Holz stated in a blog post that this model is a step toward creating AI models capable of “real-time open-world simulations.” Following video models, the company plans to develop AI models for 3D renderings and real-time AI models.
V1’s release positions Midjourney alongside competitors such as OpenAI’s Sora, Runway’s Gen 4, Adobe’s Firefly, and Google’s Veo 3, though Midjourney’s AI models are specifically geared towards creative uses.
Video: Midjourney
The V1 launch follows a lawsuit filed a week prior by Disney and Universal, alleging that Midjourney’s AI image models depict copyrighted characters, like Homer Simpson and Darth Vader.
Finally Midjourney V7 is live but it’s missing crucial features
According to Midjourney, subscribers to the $60-a-month Pro plan and $120-a-month Mega plan will have unlimited video generations in the company’s slower, “Relax” mode, while the cheapest way to try out V1 is by subscribing to Midjourney’s $10-per-month Basic plan. The company plans to reassess its video model pricing in the next month.
V1 offers custom settings to control video output. Users can select automatic animation for random movement or manual settings to describe animations in text and can also adjust camera and subject movement by selecting “low motion” or “high motion” settings. While the initial videos are five seconds, users can extend them in four-second increments up to four times, creating videos up to 21 seconds long.