Runway, an artificial intelligence startup, has unveiled its new video generation model, Gen-4, which the company claims is one of the most accurate to date, writes IZ with reference to Techcrunch. The Gen-4 model is already available to both individual and corporate users. Runway claims that its new product is capable of creating consistent characters, locations, and objects across scenes, maintaining “coherent” world environments and regenerating elements from different perspectives and positions within a single scene.
Today we're introducing Gen-4, our new series of state-of-the-art AI models for media generation and world consistency. Gen-4 is a significant step forward for fidelity, dynamic motion and controllability in generative media.
Gen-4 Image-to-Video is rolling out today to all paid… pic.twitter.com/VKnY5pWC8X
— Runway (@runwayml) March 31, 2025
“ Gen-4 can use visual references, combined with instructions, to create new images and videos that maintain consistent style, objects, locations, and more ,” Runway noted on its blog. The company claims this is achieved without the need for additional setup or additional training.
Runway has serious competitors in the video generation industry, including OpenAI and Google, but the company is actively looking for ways to differentiate itself, having signed a deal with one of the Hollywood studios and planning to invest millions of dollars in financing films created using AI-generated videos.
The Gen-4 model allows users to generate persistent characters by referencing images of those characters under different lighting conditions. To create a scene, users can provide images of the characters and describe the composition of the shot they want to create.
Gen-4 sets a new standard for video generation and is a marked improvement over Gen-3 Alpha. It excels in its ability to generate highly dynamic videos with realistic motion as well as subject, object and style consistency with superior prompt adherence and best-in-class world… pic.twitter.com/w9ACO5boJ7
— Runway (@runwayml) March 31, 2025
“ Gen-4 is distinguished by its ability to generate dynamic videos with realistic movements, as well as maintain a constant correspondence of style, objects and characters with high accuracy to instructions and understanding of the world ,” the company’s blog says.
Like all video generation models, Gen-4 was trained on a large number of video examples to “learn” the patterns of those videos and generate synthetic video. However, Runway does not disclose the sources of the training data for fear of losing a competitive advantage. At the same time, this issue could lead to legal problems, as the company is already facing a lawsuit from artists who accuse the company of using copyrighted works without permission to train its models.
Runway is also looking to raise a new round of funding that could value the company at $4 billion. According to The Information, Runway is aiming to reach $300 million in annual revenue this year after launching products such as an API for its video generation models.
While the lawsuits against Runway may drag on, AI-powered video generation tools are already posing a serious threat to the film and television industries. A 2024 study of Hollywood animators found that 75% of film studios that have implemented AI have cut, merged, or eliminated jobs altogether. It is predicted that by 2026, more than 100,000 jobs in the U.S. will be displaced or eliminated due to the introduction of generative AI in the entertainment industry.
As a reminder, we previously wrote about the main types and features of equipment for water treatment in laboratories.