MiMo-Embodied: Technisch Rapport over het X-Embodied Foundation Model
MiMo-Embodied: X-Embodied Foundation Model Technical Report
November 20, 2025
Auteurs: Xiaoshuai Hao, Lei Zhou, Zhijian Huang, Zhiwen Hou, Yingbo Tang, Lingfeng Zhang, Guang Li, Zheng Lu, Shuhuai Ren, Xianhui Meng, Yuchen Zhang, Jing Wu, Jinghui Lu, Chenxu Dang, Jiayi Guan, Jianhua Wu, Zhiyi Hou, Hanbing Li, Shumeng Xia, Mingliang Zhou, Yinan Zheng, Zihao Yue, Shuhao Gu, Hao Tian, Yuannan Shen, Jianwei Cui, Wen Zhang, Shaoqing Xu, Bing Wang, Haiyang Sun, Zeyu Zhu, Yuncheng Jiang, Zibin Guo, Chuhong Gong, Chaofan Zhang, Wenbo Ding, Kun Ma, Guang Chen, Rui Cai, Diyun Xiang, Heng Qu, Fuli Luo, Hangjun Ye, Long Chen
cs.AI
Samenvatting
Wij maken MiMo-Embodied open-source, het eerste cross-embodied foundation model dat met succes integreert en state-of-the-art prestaties behaalt in zowel Autonoom Rijden als Embodied AI. MiMo-Embodied vestigt nieuwe records op 17 embodied AI benchmarks voor Taakplanning, Affordantievoorspelling en Ruimtelijk Inzicht, en presteert eveneens uitstekend op 12 autonome rijsimulatie benchmarks voor Omgevingsperceptie, Statusvoorspelling en Rijplanning. Voor deze taken overtreft MiMo-Embodied bestaande open-source, closed-source en gespecialiseerde baseline-modellen aanzienlijk. Onze resultaten tonen aan dat deze twee domeinen, dankzij meerfasig leren, zorgvuldige dataconstructie en CoT/RL-finetuning, een sterke positieve transfer vertonen en elkaar wederzijds versterken. Wij bieden een gedetailleerde analyse van onze modelontwerpen en trainingsmethodologieën om verder onderzoek te vergemakkelijken. Code en modellen zijn beschikbaar op https://github.com/XiaomiMiMo/MiMo-Embodied.
English
We open-source MiMo-Embodied, the first cross-embodied foundation model to successfully integrate and achieve state-of-the-art performance in both Autonomous Driving and Embodied AI. MiMo-Embodied sets new records across 17 embodied AI benchmarks in Task Planning, Affordance Prediction and Spatial Understanding, while also excelling in 12 autonomous driving benchmarks across Environmental Perception, Status Prediction, and Driving Planning. Across these tasks, MiMo-Embodied significantly outperforms existing open-source, closed-source, and specialized baselines. Our results indicate that through multi-stage learning, curated data construction, and CoT/RL fine-tuning, these two domains exhibit strong positive transfer and mutually reinforce one another. We provide a detailed analysis of our model design and training methodologies to facilitate further research. Code and models are available at https://github.com/XiaomiMiMo/MiMo-Embodied.