We present InSpatio-WorldFM, an open-source real-time frame model for spatial intelligence. Unlike video-based world models that rely on sequential frame generation and incur substantial latency due to window-level processing, InSpatio-WorldFM adopts a frame-based paradigm that generates each frame independently, enabling low-latency real-time spatial inference. By enforcing multi-view spatial consistency through explicit 3D anchors and implicit spatial memory, the model preserves global scene geometry while maintaining fine-grained visual details across viewpoint changes. We further introduce a progressive three-stage training pipeline that transforms a pretrained image diffusion model into a controllable frame model and finally into a real-time generator through few-step distillation. Experimental results show that InSpatio-WorldFM achieves strong multi-view consistency while supporting interactive exploration on consumer-grade GPUs, providing an efficient alternative to traditional video-based world models for real-time world simulation.
@article{arxiv.2603.11911,
title = {InSpatio-WorldFM: An Open-Source Real-Time Generative Frame Model},
author = {InSpatio Team and Donghui Shen and Guofeng Zhang and Haomin Liu and Haoyu Ji and Jialin Liu and Jing Guo and Nan Wang and Siji Pan and Weihong Pan and Weijian Xie and Xiaojun Xiang and Xiaoyu Zhang and Xianbin Liu and Yifu Wang and Yipeng Chen and Zhewen Le and Zhichao Ye and Ziqiang Zhao},
journal= {arXiv preprint arXiv:2603.11911},
year = {2026}
}