← all releases
Waymo·Feb 6, 2026·Not specified·open source
Waymo World Model
The Waymo World Model is a generative model developed by Waymo to create hyper-realistic autonomous driving simulations. Built upon Google DeepMind's Genie 3, it generates photorealistic, controllable, multi-sensor driving scenes at scale, enabling the Waymo Driver to train and evaluate on billions of additional miles in virtual environments.
CVLLMRobotics
overview
The Waymo World Model is a generative model developed by Waymo to create hyper-realistic autonomous driving simulations. Built upon Google DeepMind's Genie 3, it generates photorealistic, controllable, multi-sensor driving scenes at scale, enabling the Waymo Driver to train and evaluate on billions of additional miles in virtual environments.
key features
- 01Generates high-fidelity camera images and lidar point clouds
- 02Simulates rare, safety-critical 'long-tail' events
- 03Offers high controllability through driving action, scene layout, and language control
- 04Built upon Google DeepMind's Genie 3 model
- 05Enhances the Waymo Driver's ability to handle diverse driving scenarios
related productsbrowse all →
Z-Imageopen sourceZ-Image is an efficient image generation foundation model developed by Tongyi-MAI, designed to produce high-quality, diverse, and stylistically versatile images. It serves as a robust base for creators, researchers, and developers seeking advanced image generation capabilities.Protenixopen sourceProtenix is an open-source, trainable PyTorch implementation of AlphaFold 3, designed for high-accuracy biomolecular structure prediction. It aims to advance accessible and extensible research tools for the computational biology community. ([github.com](https://github.com/bytedance/Protenix?utm_source=openai))