# Runway GWM-1

- Date: 2025-12-11
- Type: model
- Credit: Runway
- URL: https://aifilmhistory.org/entry/runway-gwm-1

Runway announced GWM-1, its first general world model, built on Gen-4.5 and generating frame by frame in response to the camera, actions and speech, at 720p and 24 frames per second for up to two minutes. It came as three separate models: GWM Worlds turns a still scene into a space you can explore, GWM Avatars makes conversational characters with lip-synced speech, and GWM Robotics simulates video rollouts for training robots. Worlds opened by request; Runway said the three would one day become one base model.

The same week Runway added audio and multi-shot editing to Gen-4.5. The company that made video generation a filmmaking tool was now betting that video models become simulators, the line it carried into [Solaris](https://aifilmhistory.org/entry/runway-solaris) and GWM Worlds 2.

Links:
- Runway: introducing GWM-1: https://runway.com/research/introducing-runway-gwm-1
- The Decoder: Runway unveils its first general world model: https://the-decoder.com/runway-unveils-first-general-world-model-alongside-major-gen-4-5-upgrades/
- The Batch: GWM-1 and consistent physics: https://www.deeplearning.ai/the-batch/runways-gwm-1-models-generate-videos-with-consistent-physics-for-robots-and-entertainment

---
Source: https://aifilmhistory.org/entry/runway-gwm-1
Era: 2025 · The Studio Era
Last updated: 2026-10-17