MaKe | MaHui Member Galaxy Universal Raises 700 Million RMB in Angel Round

Galbot recently announced the completion of its angel round totaling RMB 700 million, with investors including strategic and industry backers such as Meituan Strategic Investment, BAIC Capital, SenseTime Guoxiang, and iFlytek Ventures; leading financial institutions including Source Code Capital, Qiming Venture Partners, BlueRun Ventures, Matrix Partners China, and IDG Capital; and LightSource Capital serving as the exclusive financial advisor and participating in the early-stage investment.

Following this round, Galbot will continue to focus on enhancing the generalization capabilities of its embodied foundation models, while steadily advancing the technical and hardware groundwork needed to deploy these models in real-world scenarios.

Over the past two years, generative AI has evolved at breakneck speed globally. Embodied AI, as a critical physical-world vehicle for AI and the next frontier in capability extension, can dramatically raise the cognitive ceiling of traditional robotics — a development of epochal significance. Compared with conventional robots' limited scene adaptability, narrow task execution, and prohibitive marginal development costs, embodied general-purpose robots leverage more generalized cognition, more flexible action capabilities, and more mature perception to adapt to and handle diverse tasks flexibly. They represent a superior solution for robots to fully adapt to human environments, use human tools dexterously, and solve complex human tasks. Meanwhile, as core upstream components in mechanics, locomotion, and sensing become rapidly commoditized, embodied general-purpose robots have met all decisive conditions for the leap from industrial settings to households and 2C consumer applications. An industry inflection point is imminent.

Founded in May 2023, Galbot is committed to bringing generalizable embodied foundation model robots into hundreds of millions of homes. Its core team brings proven track records in research, productization, and commercialization across AI, robotics, and intelligent hardware and software.

He Wang holds a bachelor's degree from Tsinghua University's Department of Electronic Engineering and a PhD in computer science from Stanford University. He is currently an assistant professor and doctoral advisor at Peking University's Center on Frontiers of Computing Studies, director of the PKU-Galbot Joint Lab on Embodied AI, and head of embodied foundation models at BAAI. He founded and leads Peking University's Embodied Perception and Interaction Lab (EPIC Lab), with research spanning multimodal learning, embodied foundation models, autonomous driving, and 3D visual perception — with particular focus on perception and interaction for embodied robots in complex 3D environments. A prominent frontier researcher and leading figure in China's embodied AI community, He Wang has published nearly 40 papers at top-tier conferences and journals in computer vision, robotics, and AI, and has received multiple international awards including best paper and outstanding paper honors in computer vision and robotics automation.

To date, Galbot has achieved breakthroughs across multiple critical technical links in embodied AI, including fine-tuning of top-level vision-language-image foundation models and independent development of numerous skill-execution-layer core technologies, amassing substantial technical expertise and progress to become the domestic player with the most comprehensive algorithmic technology stack in embodied AI. At the recent 2024 BAAI Conference, the company unveiled its first-generation generalizable embodied foundation model robot, Galbot G1, which already demonstrates highly generalizable recognition and grasping capabilities alongside 3D visual navigation.

Additionally, Galbot has simultaneously initiated preliminary work on hardware deployment of its algorithmic technologies, with prototype refinement and scenario testing proceeding in an orderly manner. Within this year, the company expects to see its self-developed general-purpose embodied intelligent robots gradually enter real commercial scenarios and generate significant value, while making breakthrough progress in the consumer-grade task scenarios it is prioritizing.

He Wang commented: "Beyond our deep layout in retail scenarios, Galbot has already engaged in in-depth scenario exchanges and deployment validation across industrial, logistics, and education sectors, and has established strategic partnerships with multiple enterprises and institutions. Going forward, we will further explore additional application scenarios including community eldercare and home services. We hope that robots can tangibly assist humans in numerous foundational yet tedious tasks, serving as new-quality productive forces and becoming a new growth engine for enterprise development."