Paper 2508.21112
EO-1: An Open Unified Embodied Foundation Model for General Robot Control
- Published
- Aug 2025
- Research lab
- Independent
- Citations
- 21
- GitHub
- 292 stars
01 In brief
Summary
The paper introduces EO-Robotics, comprising the EO-1 model and the EO-Data1.5M dataset, to advance general robot control.
EO-1 is a 3B-parameter unified embodied foundation model that integrates multimodal understanding and robot action generation in a single decoder-only transformer, combining autoregressive decoding for text with flow-matching denoising for continuous actions.
It is trained on 135B tokens from web multimodal data, 1.2M robot episodes, and 1.5M interleaved vision-text-action samples.
The interleaved data, built via a scalable pipeline, captures temporal and spatial reasoning to enhance open-world generalization.
Evaluations show EO-1 outperforms existing models: on RoboVQA it scores 58.5 BLEU-4, on ERQA 45.5 accuracy, and on EO-Bench 44.8 overall.
In robot control, it achieves 98.2% average success on LIBERO and 72.7% on SimplerEnv WidowX, surpassing baselines like π0 and GR00T-N1.5.
Real-world tests across 28 tasks on multiple embodiments (Franka, WidowX, Agibot G-1) yield an 86% completion score, with strong generalization in visual, action, and language variations.
The work is fully open-sourced.
02 From the paper
Abstract
The human ability to seamlessly perform multimodal reasoning and physical interaction in the open world is a core goal for general purpose embodied intelligent systems. Recent vision-language-action (VLA) models, which are co-trained on large-scale robot and visual-text data, have demonstrated notable progress in general robot control. However, they still fail to achieve human-level flexibility in interleaved reasoning and interaction. In this work, we introduce EO-Robotics, consists of EO-1 model and EO-Data1.5M dataset. EO-1 is a unified embodied foundation model that achieves superior performance in multimodal embodied reasoning and robot control through interleaved vision-text-action pre-training. The development of EO-1 is based on two key pillars: (i) a unified architecture that processes multimodal inputs indiscriminately (image, text, video, and action), and (ii) a massive, high-quality multimodal embodied reasoning dataset, EO-Data1.5M, which contains over 1.5 million samples with emphasis on interleaved vision-text-action comprehension. EO-1 is trained through synergies between auto-regressive decoding and flow matching denoising on EO-Data1.5M, enabling seamless robot action generation and multimodal embodied reasoning. Extensive experiments demonstrate the effectiveness of interleaved vision-text-action learning for open-world understanding and generalization, validated through a variety of long-horizon, dexterous manipulation tasks across multiple embodiments. This paper details the architecture of EO-1, the data construction strategy of EO-Data1.5M, and the training methodology, offering valuable insights for developing advanced embodied foundation models. Project Page: https://eo-robotics.ai/eo-1.