A single destination for timely, editor-curated robotics news from around the world.
AgiBot has unveiled its GO-3 Preview robot model, which utilizes a WRAM architecture. This model has achieved top rankings in three platforms: RoboDojo, RoboTwin, and RoboColiseum. The significance of this development lies in AgiBot's commitment to open-sourcing the GO-3 Preview by the end of 2026,...
RoboParty has introduced the RP1, a 135 cm humanoid robot featuring either 24 or 30 degrees of freedom. This release is notable for being a full-stack open platform, which includes hardware, joint modules, firmware, robot models, and training tools. The significance of the RP1 lies in its open-sour...
Qualcomm Technologies has announced its acquisition of PickNik, a robotics software company known for its stewardship of MoveIt, a leading open-source framework for robot manipulation. This acquisition aims to bolster Qualcomm's robotics software capabilities, facilitating the transition from AI mod...
Amazon Web Services (AWS) has introduced the Physical AI Toolchain, an open-source toolset designed for robot developers, following the discontinuation of its RoboMaker service over a year ago. RoboMaker, launched in 2018, provided a development environment for robot software, cloud simulation, and...
Anthropic's Claude model has successfully verified security properties in 123 open-source software libraries, revealing hundreds of bugs that had evaded detection for years despite trillions of automated tests. This project, in collaboration with Theorem, utilizes mathematical proofs applied directl...
Anyverse Dynamics has launched and open-sourced MWA-Phoenix, a 7 billion parameter latent world action model designed for robotic applications. This model utilizes only RGB input and has achieved top scores on six different robot benchmarks, including performance in real-robot testing. The introduc...
SenseTime has announced the open-source release of the SenseNova-RoboRSI evaluation harness along with a 27 billion parameter embodied AI model. This development follows a comprehensive study of the GPT-6-Astra, which was tested across 12 simulated benchmarks and 8 real robots. The significance of...
RoboParty has officially launched the RP1, a full-stack open-source humanoid robot aimed at enhancing robotics research and embodied AI development. The RP1 allows researchers to access both hardware and software, facilitating customization of the system and integration of their own algorithms for t...
RoboParty has introduced the RP1, a full-stack open-source humanoid robot aimed at advancing robotics research and embodied AI development. The RP1 is available in standard and Pro versions, featuring 24 and 30 actuated degrees of freedom, respectively, and supports various modular hardware configur...
UCLA has introduced an innovative open-source project called PhoneBot, which repurposes old smartphones as the control system for a bipedal robot. This development allows smartphones that are often unused to serve as the 'brains' of the robot, enhancing its functionality and accessibility. The sign...
Asimov has announced the release of its open-source Python SDK, enabling developers to command its humanoid robot, Asimov 1. This SDK allows for high-level walking commands, direct joint control, and access to the robot's camera, microphone, and speaker. Additionally, Asimov Edge is being open-sourc...
Amazon has announced the open-sourcing of a collection of physical AI toolchains, which have been developed over more than ten years of experience in robotics. This initiative aims to enhance collaboration and innovation within the robotics community by providing access to advanced tools and framewo...
MirroS has announced the open-source release of AgentGarten, a platform where AI agents can learn by engaging in code-defined environments. Utilizing a real-time neural renderer, these agents can act, review their actions, and develop their own strategies, such as hide-and-seek tactics, during gamep...
KingKong Technology, a Beijing-based encoder and servo manufacturer, has announced the open-sourcing of its toolchain for Jumper, a crab robot featuring 22 degrees of freedom (DoF) that weighs 1.8 kg. This initiative includes simulation, reinforcement learning, and deployment capabilities, enhancing...
Amazon Web Services (AWS) has launched an open-source Physical AI Toolchain aimed at assisting roboticists in transitioning from data collection to the deployment of AI models on robots. This toolchain integrates AWS services with NVIDIA’s Physical AI software, addressing the challenge of connecting...
Amazon Web Services (AWS) has launched an open-source Physical AI Toolchain aimed at assisting robotics developers in training, simulating, deploying, and enhancing physical AI systems using AWS cloud services and Nvidia’s software stack. This toolchain encompasses synthetic data generation, model t...
Chinese robotics company RoboParty has introduced the RP1, a full-stack open-source humanoid robot aimed at advancing robotics research, education, and embodied AI development. The robot was showcased at the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026) on Sept...
RoboParty, a firm based in Pittsburgh, has introduced RP1 (ROBOTO 01), the world's first high-performance full-stack open-source bipedal humanoid robot, at IROS 2026. This robot aims to serve researchers, educators, and robotics developers by providing a customizable humanoid system that can be modi...
RoboParty has introduced the RP1 (ROBOTO 01), a bipedal humanoid robot designed for robotics researchers and developers, during its debut at IROS in Pittsburgh on September 28. The company announced on October 3 that the RP1 showcased its balance capabilities through an interactive demonstration whe...
Asimov has announced the open-source release of the training code for its humanoid robot's walking capabilities. This initiative, revealed on September 25, allows developers to inspect and modify the locomotion policy of Asimov 1, providing a foundation for adapting the robot's controller to differe...
Intrinsic, a robotics group at Google, has announced the open-source release of parts of its platform, Intrinsic Core™, aimed at enhancing physical AI applications. This platform includes capabilities such as a hardware-agnostic control framework, pose estimation, motion planning, and grasp planning...
Intrinsic, Google's industrial robotics and physical AI software company, has announced the open sourcing of foundational components of its robotics platform at ROSCon 2026 in Toronto. This initiative provides developers with access to essential technologies for robot control, motion planning, visio...
Robotiq has introduced three open-source software tools for its adaptive grippers, aimed at enhancing robotics development. These tools include a ROS 2 package, a standalone C++ SDK, and updated Isaac Sim assets, all available on Robotiq’s GitHub repository. This release marks the initial components...
Qualcomm Technologies Inc. has announced its agreement to acquire PickNik Inc., known for its MoveIt open-source manipulation framework. This acquisition aims to integrate PickNik’s MoveIt with Qualcomm's Dragonwing platform, facilitating a smoother transition for developers from AI models to real-t...
Robotiq has unveiled three open-source software tools for its adaptive grippers, including a C++ SDK, an official ROS 2 package, and updated Nvidia Isaac Sim assets. These tools are now available on GitHub and are optimized by Robotiq to enhance the functionality of their grippers in physical AI app...
Intrinsic, a subsidiary of Google, has released Intrinsic Core as open-source robotics software, enabling developers to create AI-enabled industrial robot applications from prototype to production. This software includes features such as real-time robot control, pose estimation, and motion planning,...
Intrinsic has unveiled Intrinsic Core, an open-source set of Robot Operating System-compatible capabilities aimed at simplifying robotic application development. This initiative is designed to reduce the extensive coding time currently required by robotics teams, allowing developers to leverage pre-...
NVIDIA has released Isaac ROS 5.0, which introduces advanced AI agent capabilities to the ROS developer ecosystem. This latest version also expands open source physical AI libraries and enhances deployment options across the NVIDIA Jetson platform. The significance of this release lies in its provi...
Robotiq has launched three open-source software packages aimed at enhancing its adaptive grippers. These releases include a C++ SDK, an official ROS 2 package, and updated NVIDIA Isaac Sim assets, all designed to facilitate the training of manipulation models on Robotiq components and their deployme...
Xiaomi has announced the open-source release of its MiMo-V2.6 series, which includes the MiMo-V2.6-Pro and MiMo-V2.6-Flash models. Additionally, the company has introduced MiMo-V2.6-Distill-Qwen-9B and provided research resources focused on reinforcement learning. The MiMo-V2.6-Pro achieved a score...
Light Origins has unveiled the Light-O1-Preview, a 6B model licensed under Apache-2.0. This innovative model is designed to convert human-video action priors into comprehensive robot whole-body trajectories, marking a significant advancement in robotic motion capabilities. The introduction of Light...
NVIDIA has unveiled Isaac ROS 5.0 at the ROSCon conference in Toronto, Canada, aimed at advancing robotics applications. This release introduces new agentic workflows and platform support, enabling developers to build and deploy robotics applications more efficiently. With nearly 1.3 million ROS use...
Alibaba DAMO Academy has introduced DAMO RADAR, an open-weight abdominal CT AI model, which has been published in the journal Science. This model is capable of identifying 146 findings across 18 different organs, achieving an area under the curve (AUC) of approximately 0.913, alongside expert-level...
An open-source benchmark has been developed to assess the capabilities of AI agents in engineering functional robots. This initiative comes in response to the increasing automation of software coding, where AI systems can autonomously write and revise programs. The significance of this benchmark li...
Light Origins has announced the open-sourcing of LightNav-0, a navigation model based on Qwen3-VL-4B. This model was trained using the Real2Sim2Real methodology, incorporating over 2,000 scenes and more than 4,000 hours of Visual Language Annotation (VLA) data. It excels in leading 10 monocular navi...
Xiaomi has open-sourced the Xiaomi-Robotics-U0, an autoregressive embodied world foundation model featuring approximately 4 billion parameters and full-scale weight lines of around 38 billion. This release includes training and inference tools designed to enhance robotic applications. The significa...
Shangpin Home, in collaboration with Tangyuan Technology, has launched the WorldSimReady-Home simulation dataset aimed at addressing the challenges of robotic training in complex home environments. This open-source dataset includes 100,000 square meters of high-fidelity home scenes, 10,000 interacti...
A prominent children's hospital has adopted open source NVIDIA AI technology to improve its cardiac care services. This integration allows for advanced data processing and analysis, enhancing the hospital's ability to provide accurate diagnoses and treatment plans for pediatric patients. The use of...
AGIBOT has officially released Genie Envisioner Act 2.0 (GE-Act 2.0), an open-source World-Action Model designed for training generalist manipulation policies. This release comes just after Unitree Robotics open-sourced its UnifoLM-WLA-1.0, marking a significant shift in the humanoid robotics landsc...
ACE Robotics and NTU S-Lab have announced the open-source release of Puffin-World, a multimodal world model that integrates physics, geometry, and appearance. This model utilizes the Puffin-16M dataset and achieves state-of-the-art camera absolute pose errors, reporting sub-degree accuracy across fo...
The shipment of humanoid robots is surging, with IDC projecting around 18,000 units globally by 2025, a 508% increase. However, the penetration of these robots into households remains minimal, as most are still used in entertainment, education, and data collection. Ecovacs Group Chairman Qian Dongqi...
NVIDIA has announced its acquisition of Hugging Face for $12.93 billion, marking a significant milestone in artificial intelligence history. This acquisition combines NVIDIA's leadership in AI computing with Hugging Face's extensive platform, which supports over 18 million developers and 3 million m...
Hugging Face has introduced MicroDuck, an innovative open-source robot capable of walking, talking, and roller-skating. Co-founder Thomas Wolf highlighted the increasing demand for robotics and open-source AI, noting that the company is selling approximately one MicroDuck every four seconds. The la...
Hugging Face introduced the Microduck, a $399 open-source duck robot, on Thursday, with shipping expected before Christmas. CEO Clem Delangue described the Microduck as a robot that can be taught new tricks through reinforcement learning, featuring capabilities such as waddling, picking up objects,...
NVIDIA is highlighting the contributions of partners and open source communities to local AI development throughout August. The company is showcasing its latest open models, software, and developer tools that facilitate the creation and customization of intelligent agents. New models like GLM 5.2 an...
A team from Mila and École Polytechnique de Montréal has developed an open-source robot dog named Milo, which costs approximately $2,000. This innovative solution addresses the high costs and long wait times associated with traditional guide dogs, which can exceed $50,000 and require 2-4 years for m...
Xiaomi has announced the open-source release of its embodied-AI foundation model, Xiaomi-Robotics-1, on August 5. This release encompasses the entire process from real-robot post-training to model deployment, along with code for benchmark evaluations. The model was pretrained on over 100,000 hours o...
RoboParty has successfully raised nearly 500 million yuan through angel and Pre-A funding rounds, with CATL exclusively participating in the Pre-A round. The startup, founded by 22-year-old Huang Yi from Harbin Institute of Technology, focuses on developing an open-source embodied platform featuring...
Digua Robot has introduced the X-Lens depth estimation model, which features 40 million parameters, significantly fewer than traditional models. This innovative solution can process mixed inputs from both fisheye and pinhole cameras, producing dense depth maps with real-world scale. On the Digua Xur...
Tencent has announced the open-source release of three embodied foundation models during the WAIC 2026 event. These models include a Visual Language Model (VLM) designed for scene understanding, the RxBrain cognitive model that facilitates planning with visual states, and the VLA model, which suppor...
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.
