G1
PulseAugur coverage of G1 — every cluster mentioning G1 across labs, papers, and developer communities, ranked by signal.
- 2026-10-25 research_milestone ETH Zurich demonstrated that the G1 robot can learn complex physical tasks like ladder climbing with only ten minutes of human training data. source
- 2026-09-09 product_launch Unitree Robotics demonstrated its G1 humanoid robot sparring autonomously using the UnifoLM-X2-1.0 world model. source
- 2026-08-26 product_launch A demonstration video revealed the G1 embodied AI model performing complex tasks autonomously with Unitree Robotics hardware. source
8 day(s) with sentiment data
-
XGIMI launches AI glasses and wearables with long-term memory features
XGIMI has unveiled its new lineup of AI-powered wearable devices, including the MemoMind One AI display glasses, G1 AI memory glasses, and L1 AI memory bracelet. The MemoMind One focuses on integrating AI into daily lif…
-
New CBFI framework aims to reveal hidden interactions in ML models
Researchers have developed a new framework called localized case-based feature importance (CBFI) to better understand the decision-making processes of complex machine learning models. Unlike existing methods like SHAP a…
-
New benchmark assesses AI world models' understanding of robotic policy failures
Researchers have introduced WorldFailBench, a new benchmark designed to evaluate how well world models understand policy failures in robotics. The benchmark aims to determine if these models accurately simulate the reas…
-
Chinese AI portal WEILAI.WANG tracks frontier models and robotics
WEILAI.WANG is a Chinese-language portal designed to track the rapid advancements in artificial intelligence, including new models, coding agents, robotics, and funding. The site aims to provide insights into what these…
-
ETH Zurich trains robot G1 to climb ladders with 10 minutes of data
Researchers at ETH Zurich have developed a new method for teaching robots complex physical tasks, such as climbing ladders and navigating challenging terrain. This approach utilizes contact prediction and requires only …
-
Humanoid robots learn table tennis; NEURA Robotics releases 'robot brain' foundation model
Researchers have developed a framework enabling two humanoid robots to play table tennis through coordinated AI-driven movements. This system, a collaboration involving NVIDIA, the University of Hong Kong, and others, t…
-
Even Realities G2 smart glasses improve display but complicate controls
Even Realities' G2 smart glasses offer improved display technology, including higher resolution, a wider field of view, and increased brightness compared to their predecessor. The new model also boasts enhanced durabili…
-
Unitree Robotics launches UniUni swarm control software for H2 and G1 humanoids
Unitree Robotics has launched UniUni, a new software designed for controlling swarms of their H2 and G1 humanoid robots. This application, available for Windows and macOS, allows users to choreograph robot movements, sy…
-
New TEMPO method enables real-time robot policy adaptation
Researchers have developed TEMPO, a novel method for test-time adaptation (TTA) of robot locomotion policies. This approach allows robots to adapt their behavior in real-time without needing explicit rewards or extensiv…
-
Robots learn tennis from imperfect human data using latent action space · 1 source tracked
Researchers from Galaxy General and Tsinghua University have developed a novel approach called LATENT to enable humanoid robots to learn tennis skills using imperfect human motion data. Instead of relying on perfect, co…
-
OpenAI pledges daily improvements or quota resets for Codex/Work users
OpenAI's Head of Codex, Tibo, has committed to delivering a tangible improvement to Codex/Work users daily for the next 28 days, or resetting their usage quotas. This initiative stems from user feedback highlighting iss…
-
New OTRetarget method enables robots to mimic human motion and object manipulation
Researchers have developed OTRetarget, a novel method for transferring human motion to humanoid robots, including the manipulation of objects. This approach uses entropic optimal transport to maintain consistency in int…
-
Unitree G1 robot spars autonomously using UnifoLM-X2-1.0 world model
Unitree Robotics has demonstrated its G1 humanoid robot sparring with a trainer, utilizing the UnifoLM-X2-1.0 world model. This advanced model enables the robot to predict physical states and plan its movements in real-…
-
Embodied AI G1 shows cross-robot collaboration and tool use · 1 source tracked
A demonstration video has surfaced showcasing a novel embodied AI model capable of complex, autonomous tasks using two distinct robot platforms, Unitree Robotics and Unitree Robotics. The AI, referred to as G1, exhibits…
-
Chinese startup Super-Dimension Dynamics unveils full-stack embodied AI system
Chinese embodied AI startup Super-Dimension Dynamics has unveiled a full-stack solution for developing humanoid robots, including their KAI Bot humanoid robot, KAI Hand dexterous manipulator, KAI Halo data collection de…
-
New frameworks enhance embodied AI simulation with cross-embodiment learning and procedural generation
Researchers have developed new frameworks for creating more realistic and versatile simulation environments for embodied AI. CLAP enables training on diverse video data across different robotic and human agents by recon…
-
Humanoid robots play autonomous table tennis at World Robot Games · 8 sources tracked
At the 2026 World Robot Games, a team from the University of Hong Kong and KAI research demonstrated the first fully autonomous humanoid robot table tennis match. Two robots, utilizing the SMASH 2.0 system, played an en…
-
Robots learn complex tasks using simulated expert data and sparse rewards
Researchers have developed a novel approach to train robots for complex locomotion and manipulation tasks by leveraging Sample-based Model Predictive Control (SMPC) in simulation. This method generates large datasets th…
-
AI glasses startup Yiwen achieves unicorn status amid intensifying competition
Yiwen, an AI glasses startup, has achieved unicorn status with a reported revenue growth of over 100% in the second quarter and aims for 1 billion RMB in revenue within the next year. The company has secured 1 billion R…
-
Chinese firms dominate humanoid robot market in H1 2026, AgiBot leads
In the first half of 2026, Chinese companies dominated the global humanoid robot market, capturing 97% of shipments. AgiBot emerged as the leading vendor, surpassing Unitree Robotics by securing 44% of the market share …