Amap, Alibaba's location-based services platform, released a pair of generative AI models, ABot-World-0 and ABot-3DWorld-0, that can produce interactive video and three-dimensional scenes in a unified framework, with the ability to run continuously for more than an hour on a single consumer-grade graphics processor. The models are designed to overcome key limitations in "world models," as existing video-generation models typically degrade after about one minute of inference. The ABot-World model integrates scene navigation and character control into a single training objective and applies a temporal-consistency algorithm, allowing stable generation for over an hour without quality drop. The ABot-3DWorld-0 model shares the same inference pathway as Amap's previously released ABot-Earth0.5, enabling a single architecture to produce 3D Gaussian Splatting assets across indoor, street-level, and aerial city scales. Amap integrates both models into a single product and introduces a "spatial teleport" mechanism for creating anchor points between generated 3D worlds. All inference tasks can run locally on a single consumer-grade GPU, and model weights, inference code, and deployment guides have been partially open-sourced.