Runway

Runway Gen 4 (https://runwayml.com) is, as of June 2025, one of the most highly-respected AI video generator models, on a par with Kling 2.0 but not as good as Veo 3 at generating life-like video with sound. It's certainly more affordable than Veo 3 and I signed up for a one month free trial. 

To quote from the video below (from March 2025): Gen 4 sets a completely new standard for both image and video generation and it's the first model ever that we're putting out that achieves world consistency. That means that you can create consistent worlds with consistent environments, objects, location and characters. And when you can do that you can start to tell longer form narrative content with actual continuity. You can generate the same characters, the same objects, the same locations across different scenarios so you can block your scenes and tell your stories with intention over and over and over again.

Here below is a 2 minute video made with Runway Gen 4 followed by a 24 minute video explaining how it was made. The explanation focuses on the use of reference images to allow for character consistency between scenes, a feature that Veo 3 does not have. I was attracted to this video because I had been meaning to delete my 23andme DNA data - this video was the trigger that finally made me do it.

Although I have chosen to use Veo 3 for my video project, the other AI companies have not stood still. In August 2025 Runway introduced Aleph, which looks interesting, judging by this video.