Today, we’re launching D1, a desktop embodied AI platform designed to support teaching and application development at universities and vocational colleges. D1 combines a 6-DoF robotic arm, an 11-DoF dexterous hand with 6 active and 5 passive degrees of freedom, and a 2-DoF active vision head, bringing 19 degrees of freedom together in one compact system. Built for hands-on learning, D1 offers: - Compatibility with leading simulators, including Isaac Gym, Isaac Sim, MuJoCo, and PyBullet - Teleoperation via exoskeleton, keyboard, or VR - A unified Python API and Ubuntu support - Support for open-source pretrained models, including Being-H, ACT, and π-series models From teleoperation to real-world robot testing, students can experience the full embodied AI development workflow. At BeingBeyond, we build general-purpose embodied AI foundation models. With D1, we bring that expertise into education by turning our experience in model development, data processing, and hardware and software integration into practical tools and course resources. In this way, D1 helps schools prepare students for the needs of the industry. Interested in partnering with us? 📩 [email protected]

Today, we introduce Being-H0.8, a Latent Tactile World Action Model. For the first time, Being-H0.8 brings touch into large-scale embodied foundation model pretraining. It unifies vision, touch, action, and future world dynamics within a shared latent space, enabling robots to understand what they see, what they feel, what they do, and how their actions change the world. Built on more than 500,000 hours of human egocentric video collected in collaboration with a broad network of data partners, Being-H0.8 marks a new step toward embodied foundation models that can perceive, interact, predict, generate actions, and generalize across robot embodiments. With an unprecedented learning paradigm, BeingBeyond is shaping the future of general-purpose embodied foundation models. Blog: research.beingbeyond.com/bei…

We believe human video is the most important path for robots to understand and interact with the physical world. Today we are introducing Being-H0.7, BeingBeyond’s third-generation flagship model. Being-H0.7 advances a new paradigm for world models built on latent-space reasoning, and scales pretraining to 200,000 hours of egocentric video. Check more about our work👇 Blog: research.beingbeyond.com/bei… Paper: research.beingbeyond.com/pro…

Introducing BeingBeyond U1, the world’s first Real DexUMI. U1 brings embodiment-agnostic dexterous hand data collection to real-world manipulation, taking a major step toward general-purpose dexterous manipulation models. From data collection to transfer, deployment, and execution, U1 pushes UMI beyond the gripper era and into the age of dexterous hands.

Big month at BeingBeyond 🎉 We’re excited to share that in the past month we’ve had 8 papers accepted across top conferences and journals: CVPR’26 ×5, ICLR’26 ×1, ICRA’26 ×1, RAL ×1. Check our papers below👇 [CVPR'26] DemoFunGrasp: Universal Dexterous Functional Grasping via Demonstration-Editing Reinforcement Learning research.beingbeyond.com/dem… arxiv.org/pdf/2512.13380 [CVPR'26] End-to-End Language-Action Model for Humanoid Whole Body Control arxiv.org/pdf/2511.19236 [CVPR'26] Joint-Aligned Latent Action: Towards Scalable VLA Pretraining in the Wild (To appear) [CVPR'26] OpenT2M: No-frill Motion Generation with Open-source, Large-scale, High-quality Data (To appear) [CVPR'26] Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos research.beingbeyond.com/vip… arxiv.org/pdf/2512.13080 [ICLR'26] DemoGrasp: Universal Dexterous Grasping from a Single Demonstration research.beingbeyond.com/dem… arxiv.org/pdf/2509.22149 [ICRA'26] Towards Proprioception-Aware Embodied Planning for Dual-Arm Humanoid Robots arxiv.org/pdf/2510.07882 [RAL] DemoHLM: From One Demonstration to Generalizable Humanoid Loco-Manipulation research.beingbeyond.com/dem… arxiv.org/pdf/2510.11258 We’ll keep focusing on VLA models, Dexterous Manipulation, and Whole-Body Control. BeingBeyond looks forward to seeing everyone at the venues!🥳