AI image and video generation company Runway has launched its first model, joining a growing number of startups and big tech companies in the race to release global models. The model, called GWM-1, works through frame-by-frame predictions to create simulations that understand physics and how the world actually behaves over time, the company said.
A world model is an AI system that learns an internal simulation of how the world works, allowing it to reason, plan, and act in any real-world scenario without training.
Runway, which announced its Gen 4.5 video model earlier this month topping both Google and OpenAI on the Video Arena leaderboard, said its GWM-1 world model is “more general” than Google’s Genie-3 and other competitors. The company is pitching it as a model that can create simulations to train agents in a variety of fields, including robotics and life sciences.
Runway has released specific trends for new world models called GWM-Worlds, GWM-Robotics, and GWM-Avatars.
GWM-Worlds is a model app that allows you to create interactive projects. Users can set the scene through prompts, and as they explore the space, the model understands geometry, physics, and lighting to generate the world. Runway said that while Worlds can be useful for games, it’s also good for teaching agents how to move and act in the physical world.
With GWM-Robotics, the company aims to use synthetic data enriched with new parameters such as changing weather conditions and obstacles. According to Runway, this method could reveal when and how robots may violate policies and instructions in a variety of scenarios.
Runway also builds realistic avatars under GWM avatars to simulate human behavior. Companies like D-ID, Synthesia, Soul Machines, and even Google have worked to create human avatars that look real and function in areas such as communication and training.
tech crunch event
san francisco
|
October 13-15, 2026
In addition to releasing new global models, the company is also updating the basic 4.5 generation model released earlier this month. A new update adds native audio and long multi-shot generation capabilities to the model. The company says the model allows users to generate one-minute videos with consistent characters, native dialogue, background audio, and intricate shots from different angles.
The 4.5-gen update brings Runway closer to competitor Kling’s all-in-one video suite, which was also released earlier this month, especially when it comes to native audio and multi-shot storytelling. It also shows that video generation models are moving from prototypes to production-ready tools.
Runway’s updated 4.5 generation model will be available to enterprise customers first, and then to all paid plan users in the coming weeks.
The company said it will make GWM-Robotics available through its SDK. The company added that it is in active discussions with several robotics companies and companies about using GWM-Robotics and GWM-Avatars.
