tech
December 11, 2025
Runway releases its first world model, adds native audio to latest video model
Runway debuts a physics-aware world model that simulates reality to train agents and power video, robotics and avatar applications.
TL;DR
- Runway has launched its first world model, GWM-1, which creates simulations with an understanding of physics.
- World models learn internal simulations to reason, plan, and act without needing training on every scenario.
- GWM-1 is described as more 'general' than competitors like Google's Genie-3.
- Specific versions of GWM-1 include GWM-Worlds for interactive scene generation, GWM-Robotics for training agents with synthetic data, and GWM-Avatars for simulating human behavior.
- Runway's Gen 4.5 video model has been updated with native audio and long-form, multi-shot generation capabilities.
- The updated Gen 4.5 model can generate one-minute videos with character consistency, dialogue, background audio, and complex shots.
- GWM-Robotics will be available via an SDK, and Runway is in discussions for its use in robotics and enterprises.
- The advancements bring Runway closer to competitor Kling's video suite and indicate a shift towards production-ready video generation tools.
Continue reading
the original article