开源多模态机器人基础模型，打通感知、推理与规划闭环
RynnBrain: Open Embodied Foundation Models
- 统一框架整合空间时序理解与物理推理能力
- 在20个机器人基准上超越现有模型，30B版本表现最优
- 适合机器人研发、具身智能研究者快速构建应用
尽管多模态基础模型进展迅速，具身智能领域仍缺乏一个统一的、基于物理世界的通用基础模型，难以实现感知、推理与规划的协同。我们提出RynnBrain，一个开源的时空基础模型，用于具身智能。该模型在统一框架下强化四大核心能力：全面的自我中心理解、多样化的时空定位、物理约束的推理以及物理感知的规划。RynnBrain包含三个规模模型（2B、8B、30B-A3B MoE）和四种针对下游任务微调的变体，分别适用于导航（RynnBrain-Nav）、规划（RynnBrain-Plan）、视觉语言动作（RynnBrain-VLA）及复杂空间推理（RynnBrain-CoP）。在20个具身智能基准和8个通用视觉理解基准上的评估表明，RynnBrain显著优于现有具身基础模型。后训练模型套件进一步验证了其两大潜力：（i）支持物理驱动的推理与规划；（ii）可作为强大预训练主干，高效适配多种具身任务。
原文摘要 · Abstract (English)
Despite rapid progress in multimodal foundation models, embodied intelligence community still lacks a unified, physically grounded foundation model that integrates perception, reasoning, and planning within real-world spatial-temporal dynamics. We introduce RynnBrain, an open-source spatiotemporal foundation model for embodied intelligence. RynnBrain strengthens four core capabilities in a unified framework: comprehensive egocentric understanding, diverse spatiotemporal localization, physically grounded reasoning, and physics-aware planning. The RynnBrain family comprises three foundation model scales (2B, 8B, and 30B-A3B MoE) and four post-trained variants tailored for downstream embodied tasks (i.e., RynnBrain-Nav, RynnBrain-Plan, and RynnBrain-VLA) or complex spatial reasoning tasks (i.e., RynnBrain-CoP). In terms of extensive evaluations on 20 embodied benchmarks and 8 general vision understanding benchmarks, our RynnBrain foundation models largely outperform existing embodied foundation models by a significant margin. The post-trained model suite further substantiates two key potentials of the RynnBrain foundation model: (i) enabling physically grounded reasoning and planning, and (ii) serving as a strong pretrained backbone that can be efficiently adapted to diverse embodied tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。