WorldClaw
WorldClaw is a fully agentic, coarse-to-fine framework for generating large-scale, freely explorable, and editable 3D open worlds from open-ended text prompts. Rather than synthesizing an entire world in a single pass, it follows a global-to-regional construction strategy that maintains coherent spatial organization while adding rich local detail. Planning agents first translate a prompt into a structured scene specification covering regions, terrain, assets, materials, visual style, and spatial relationships. It builds a globally coherent terrain foundation using semantic layouts, reusable assets, generative or procedural materials, and region-aware height fields. For regions requiring greater detail, WorldClaw creates terrain-conditioned compositions, reconstructs editable textured meshes, and recovers their placement within the scene. Render-based agents further refine terrain geometry, objects, appearance, arrangements, and contacts with the environment.
Learn more about WorldClaw
Odyssey-2 Max
Odyssey-2 Max is a scaled, real-time world simulation model designed to move beyond traditional generative AI by learning how the physical world behaves and enabling continuous, interactive environments. It represents the third and most advanced model in the Odyssey-2 family, significantly increasing scale with three times the parameters and ten times the training compute compared to Odyssey-2 Pro, which unlocks new emergent behaviors and more stable, realistic simulations. It is built to simulate physics, human motion, interaction, and environmental dynamics in real time, generating continuous streams of visual output that respond instantly to user input instead of producing fixed clips. Unlike conventional video models that generate short, precomputed sequences, Odyssey-2 Max produces long-running simulations that evolve frame by frame, allowing users to interact with the environment as it unfolds.
Learn more about Odyssey-2 Max
Genie 3
Genie 3 is DeepMind’s next-generation, general-purpose world model capable of generating richly interactive 3D environments in real time at 24 frames per second and 720p resolution that remain consistent for several minutes. Prompted by text input, the system constructs dynamic virtual worlds where users (or embodied agents) can navigate and interact with natural phenomena from multiple perspectives, like first-person or isometric. A standout feature is its emergent long-horizon visual memory: Genie 3 maintains environmental consistency over extended durations, preserving off-screen elements and spatial coherence across revisits. It also supports “promptable world events,” enabling users to modify scenes, such as changing weather or introducing new objects, on the fly. Designed to support embodied agent research, Genie 3 seamlessly integrates with agents like SIMA, facilitating goal-based navigation and complex task accomplishment.
Learn more about Genie 3
Project Genie
Project Genie is an experimental AI system from Google that generates interactive worlds in real time. It allows users to create living, explorable environments using simple text or image prompts. As you move through a world, Genie dynamically builds the landscape around you, making each experience unique. Users can design characters and choose how they explore, from walking and driving to flying and riding. The platform supports a wide range of environments, including natural landscapes, fictional worlds, and scenes generated from photos or artwork. Genie reacts to movement, physics, and user actions to create a continuous sense of discovery. Project Genie showcases the future of real-time, AI-generated interactive environments.
Learn more about Project Genie