6thWave: AI News Hub

RunwayML, a pioneer in AI-driven video generation, has unveiled its latest model, Gen-3 Alpha, making it generally available to users on its platform. Unlike its predecessors, Gen-3 Alpha is a paid feature, starting at $12/month per editor when billed yearly. This new model allows users to create hyper-realistic videos from text, image, or video prompts, offering advanced features such as imaginative transitions, precise key-framing, and expressive human characters. Developed by a team of research scientists, engineers, and artists, Gen-3 Alpha marks a significant upgrade over the previous Gen-1 and Gen-2 models in terms of speed, fidelity, consistency, and motion. Initially, the model will power text-to-video generation, with plans to expand to other modes like image-to-video and video-to-video. The videos generated can be up to 10 seconds long, which is shorter than OpenAI’s promised one-minute-long videos with Sora. Gen-3 Alpha is expected to evolve over time, with a free version anticipated in the future, as part of Runway’s vision to build General World Models capable of simulating a wide range of real-world interactions.

Source.