ACE Robotics
2

ACE Robotics

Empowering Every Robots With A Super- Brain
ACE Robotics releases ACE-Brain-0.6, says it outperforms GPT-6 Astra on spatial grounding - RobotHOT

ACE Robotics releases ACE-Brain-0.6, says it outperforms GPT-6 Astra on spatial grounding

ACE Robotics has released a visual-spatial grounding-enhanced embodied foundation model, ACE-Brain-0.6, and said the 9-billion-parameter system outperformed OpenAI’s GPT-6 Astra on six of eight grounding benchmarks. The company said the model is designed to improve target identification, spatial localization and interaction anchoring—capabilities it described as critical for translating high-level task reasoning into reliable physical action. It builds on the earlier ACE-Brain-0.5 and focuses on visual grounding: identifying target objects, interaction areas and spatial references, and linking them to position, displacement and motion parameters. In seven public benchmarks, ACE-Brain-0.6 averaged 74.53 points versus 73.68 for GPT-6 Astra, according to the company. It said the model beat GPT-6 Astra on six of eight evaluations, including Where2Place (76.55 vs. 71.10), PixMo-Count (89.41 vs....

Embodied AI
ACE ROBOTICS open-sources ACE-Ego-Hand for 3D hand reconstruction from egocentric video, claims 45 benchmark wins - RobotHOT

ACE ROBOTICS open-sources ACE-Ego-Hand for 3D hand reconstruction from egocentric video, claims 45 benchmark wins

ACE ROBOTICS, in collaboration with the Chinese University of Hong Kong’s Multimedia Laboratory and Nanyang Technological University, has open-sourced ACE-Ego-Hand, a 3D dual-hand shape reconstruction model designed for first-person video, generating approximately 5,000 hours of training data that could serve as a foundational dataset for embodied AI model training. The model extracts structured 3D bimanual manipulation trajectories from massive egocentric video footage, translating human grasping, assembly and hand-coordination actions into machine-learnable data. It marks the first time a video generation model has been repurposed as a DreamHand encoder for end-to-end training, outputting hand pose, shape and 3D position in a single forward pass without frame-by-frame detection or temporal regression. Efficiency gains over traditional approaches Unlike previous hand models reliant on...

Robot Software & Simulation

Briefs

  • Autonomous first place: HKUST and GENISOM AI win with Gangbeng L2

    A team from the Hong Kong University of Science and Technology, working with GENISOM AI, won first place in the autonomous group for quadruped-wheeled robots at the 2026 Robot Warrior Challenge Autumn Competition in Beijing, the company said. GENISOM AI said its “GENISOM Gangbeng L2” robot, equipped with a robotic arm, competed in the autonomous operation category. The company said the result followed its “GENISOM Tongchui M1” platform’s six awards in the autonomous navigation category at the spring event, and showed its position in autonomous operation in complex, unstructured environments.

  • China's Guoyu Xingchen raises nearly 100 mln yuan in Series A funding

    Chinese space robotics startup Guoyu Xingchen has raised nearly 100 million yuan ($14 million) in a Series A round jointly invested by Yijing Capital, Fosun Chuangfu and Luyao Capital, the company said. The proceeds will focus on building space robot clusters and expanding commercialization, as the company seeks to accelerate China's in-orbit services from technical validation to scaled commercial delivery.

  • BYD granted design patent for embodied AI robot

    Chinese automaker BYD has been granted a design patent for an “embodied intelligent robot,” according to a patent abstract. The design covers a robot capable of walking, balancing and interacting, with potential applications in service, education and industrial settings.

  • China issues guidelines to speed tech transfer in consumer goods, including brain-computer interfaces and embodied AI

    China’s Commerce Ministry and eight other government departments have issued guidelines on accelerating technology-driven commerce, saying the country will cultivate technology transfer and commercialization institutions in the consumer goods sector. The guidelines, jointly released by the Commerce Ministry, Cyberspace Administration of China, National Development and Reform Commission, Ministry of Education, Ministry of Science and Technology, Ministry of Industry and Information Technology, Ministry of Culture and Tourism, People’s Bank of China and State Administration for Market Regulation, call for speeding the market conversion of frontier technologies including brain-computer interfaces and embodied intelligence.

  • China's DareRobot raises nine-figure yuan in Pre-A+ funding round

    Chinese industrial robotics startup DareRobot said it has raised a nine-figure yuan amount in a Pre-A+ funding round, without disclosing the exact figure. Investors included Fosun RuiZheng, Qingshi Asset Management, Zhengxi Capital, Xigao Investment, Bozhong Precision Industry and Suzhou Venture Capital, the company said. DareRobot, founded in 2022, makes intelligent cutting, grinding, welding and coating robots for heavy-industry sectors including defense, steel, marine and shipbuilding, and energy equipment. It targets automation in complex workpieces and high-mix, low-volume production. The proceeds will mainly fund research and development of core embodied-AI technologies for heavy industry, robot product upgrades and market expansion, the company said.

  • Poxiao AI, Tinno Mobile launch full-body multimodal data collection system

    Chinese embodied-AI company Poxiao AI and handset maker Tinno Mobile have jointly released SeePhys Hub, a full-body multimodal data collection system. The system comprises a four-camera head ring, wrist bands, waist and leg motion trackers and tactile gloves, the companies said. It can simultaneously collect first-person visual, two-handed tactile and full-body motion data, filling gaps in conventional ego-data collection, which struggles to capture body movement and contact pressure. A companion data-processing platform converts the collected content into multimodal training data for robot learning, they said.

  • China's Youibot clears CSRC filing for Hong Kong IPO, eyes first mobile-manipulation robot listing

    China's securities regulator on Oct. 16 published a filing notice for Hefei Youibot Robotics Co., Ltd.'s overseas listing and full circulation of domestic unlisted shares, clearing a precondition for a Hong Kong Stock Exchange hearing and putting the company on track to become the first listed mobile-manipulation robot maker. Youibot plans to issue no more than 49,477,800 overseas-listed ordinary shares and list on the Hong Kong Stock Exchange, according to the notice. Thirty-seven shareholders plan to convert a combined 243,804,260 domestic unlisted shares into overseas-listed shares, achieving "full circulation."

  • AGIBOT, Ant Afu deploy smart triage robot at Peking University People's Hospital

    The smart triage project jointly developed by AGIBOT and Ant Afu has been deployed at Peking University People's Hospital, with Lingxi X2 serving as a "silicon-based receptionist" in outpatient services. The robot can provide patients and visitors with department guidance, consultation process information, answers to medical insurance policy questions and on-site navigation. By connecting to the hospital's private knowledge base, Lingxi X2 can also deliver standardized medical guidance based on in-hospital information. Over 66 working days, it completed more than 46,000 interactions, averaging about 707 per day.
MORE

Hot Topics

RobotHOT - China & Global Robotics Insights

RobotHOT brings China & Global Robotics Insights. Explore robotics, embodied‑intelligence updates, startup news and industry analysis for professionals worldwide.

About USContact US