To use the new feature, users can navigate to Runway’s website and click on the “try Gen 3-Alpha” language. They will then be able to upload still images and enter text prompts to guide the AI video generation. Users can choose to generate videos that are either 5 or 10 seconds long, with the latter requiring 40 credits through Runway’s pay-to-play and subscription tiers.
Tests of Runway’s Gen-3 Alpha image-to-video update have shown that it is surprisingly fast while maintaining high quality. The feature automatically attempts to detect and block the creation of videos from explicit still imagery or well-known figures, such as politicians.
Runway initially released the feature online briefly for some subscribers over the weekend before temporarily removing it and then putting it back up. The company also showcased 10 interesting videos generated from still images with the Gen-3 Alpha model on its X account.
According to Runway’s co-founder and CEO Cristóbal Valenzuela, the image-to-video update is a major improvement that enhances the artistic control and consistency of video generations. Valenzuela introduced the new feature on his personal account with the phrase “it’s time” followed by his own image-to-video creation.
Runway is one of several companies competing to offer fast and high-quality generative video models, including OpenAI, Kuaishou Technology, Luma AI, and Pika. While OpenAI’s model is not yet available to the public, the other models can be accessed and used by users. With Runway’s still image-to-video update, the field of filmmaking and video creation is being revolutionized.
However, Runway and other AI companies are currently facing class-action lawsuits from creators who claim that the practice of scraping and training on publicly posted and copyrighted material without permission or compensation is a violation of copyright law. The outcome of these lawsuits will have a significant impact on the future of AI video and creative tools.
