Runway unveiled its “better and faster” video model with Gen-3 AI two weeks ago, promising access “in the coming days.” It’s now available to a select group of Creative Partners members, who have shared incredible insights into what possible.
Gen-3 is a “Sora-like” model, offering more natural motion, improved realism, and 10-11 second initial video clips compared to the previous generation’s 2-3 seconds.
The new model has been rebuilt from the ground up using next-generation infrastructure that has been purpose-built for large-scale multimodal training.
This new model was trained on both image and video for better realism, however it only works with text to video conversion during testing.
Videos we’ve shared include a spaceman in a music video, a hyper-realistic woman aging as the camera moves around her, and title screens with animated words.
How well does Gen-3 work?
I’m not a Partner, so I can’t provide first-hand experience using the Gen-3, but from what those with access have said, it needs image-to-video transfer, but otherwise it’s impressive.
The thing I’m most excited about is the fact that Gen-3 even works well over lip sync. This allows full body animation along with precise lip movement.
1. Spaceman in a music video
The Gen-3 Alpha track is INCREDIBLE! 🤖As a creative partner I only had early access for just over 24 hours and here are some fun visuals I created 🧵⬇️#gen3 #Runway 🔊🆙 pic.twitter.com/ZR9GhSUn4TJune 30, 2024
Creative Partner NoBanksNearby made an impressive music video in the style of a spaceman dance sequence with an impressive amount of bass.
In the same thread, they also shared a number of other spaceship-inspired videos, all made with the Runway Gen-3. Writing on X: “I’m using my creativity and having a lot of fun and we’re not even at full version yet with [image-to-video]”.
2. Title screen animation
Wow…Runway Gen-3 excels at creating amazing title screen animations. 9 impressive examples:1. pic.twitter.com/PAi80T7WNAJune 30, 2024
The ability to produce legible and consistent text is something that AI image tools are still getting to grips with, but it looks like Runway has it in Gen-3.
ProperPrompter has shared a number of impressive clips that show the credits animations, including one that shows the word Runway emerging from the surrounding leaves.
3. VFX and cinematic moments
5. “Astronaut on a Scary Spaceship Science Fiction Movie HR Giger” pic.twitter.com/FFlt4SR0ejJune 29, 2024
Javi Lopez, founder of Magnific, an AI upscaler & enhancer, created some impressive graphic footage including “film footage of water inside a fire in a fast spiral” and “film footage of a gorgeous woman from water”.
My favorite has to be the astronaut walking through a scary spaceship that gives off both VFX and AAA game vibes at the same time.
4. Zombies at the table
User Maxescu asked the public for challenges to share, and one of them was a zombie in a tuxedo sitting at a table in a team drinking 1940s ocean liner.
The exact prompt was: “Black and white footage of a zombie in a tuxedo sitting at a table on a 1940s ocean liner drinking tea,” Jason Wissinger shared, along with his own Luma Dream Machine clip.
5. Monster from the deep
“film footage of a hideous monster rising from the River Thames in London” Gen 3 pic.twitter.com/X31GQLOSL7June 28, 2024
Uncanny Harry has created a stunning video of a “hideous monster rising from the River Thames in London”. It shows how artificial intelligence will enable even small indie films to include blockbuster-style VFX.