HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents

TL;DR

HY-Embodied-0.5 introduces specialized foundation models for embodied agents, enhancing visual perception and reasoning with a Mixture-of-Transformers architecture and on-policy distillation, validated across numerous benchmarks and real-world robot tasks.

Contribution

The paper presents a novel family of models tailored for embodied intelligence, combining efficient and large variants with advanced architecture and training paradigms, and open-sourcing the code.

Findings

The MoT-2B model outperforms similar-sized state-of-the-art models on 16 benchmarks.

The 32B model achieves performance comparable to frontier models like Gemini 3.0 Pro.

The models demonstrate effective real-world robot control capabilities.

Abstract

We introduce HY-Embodied-0.5, a family of foundation models specifically designed for real-world embodied agents. To bridge the gap between general Vision-Language Models (VLMs) and the demands of embodied agents, our models are developed to enhance the core capabilities required by embodied intelligence: spatial and temporal visual perception, alongside advanced embodied reasoning for prediction, interaction, and planning. The HY-Embodied-0.5 suite comprises two primary variants: an efficient model with 2B activated parameters designed for edge deployment, and a powerful model with 32B activated parameters targeted for complex reasoning. To support the fine-grained visual perception essential for embodied tasks, we adopt a Mixture-of-Transformers (MoT) architecture to enable modality-specific computing. By incorporating latent tokens, this design effectively enhances the perceptual representation of the models. To improve reasoning capabilities, we introduce an iterative, self-evolving post-training paradigm. Furthermore, we employ on-policy distillation to transfer the advanced capabilities of the large model to the smaller variant, thereby maximizing the performance potential of the compact model. Extensive evaluations across 22 benchmarks, spanning visual perception, spatial reasoning, and embodied understanding, demonstrate the effectiveness of our approach. Our MoT-2B model outperforms similarly sized state-of-the-art models on 16 benchmarks, while the 32B variant achieves performance comparable to frontier models such as Gemini 3.0 Pro. In downstream robot control experiments, we leverage our robust VLM foundation to train an effective Vision-Language-Action (VLA) model, achieving compelling results in real-world physical evaluations. Code and models are open-sourced at https://github.com/Tencent-Hunyuan/HY-Embodied.