Group 1: Financial Performance - In the first half of 2025, the company's product business revenue increased by 22.66% year-on-year, with a gross margin of 38.83%, showing slight growth compared to the previous year [2] - The industrial robot business revenue grew by 22.55% year-on-year, with self-produced multi-joint industrial robots seeing an impressive revenue increase of 80.86% [2] - The revenue from injection molding machines, supporting equipment, and automatic feeding systems decreased by 0.87%, while the revenue from supporting equipment and automatic feeding systems increased by 29.64% [3] Group 2: Product Development - The company's first humanoid robot product, "Xiao Tuo," is capable of autonomous reasoning and decision-making for complex tasks, demonstrating the potential of humanoid robots in industrial applications [4] - "Xiao Tuo" integrates a 3D camera and AI visual inspection system to identify the quality of injection molded parts in real-time, achieving full-process automation from inspection to operation [4] Group 3: Impact on CNC Machine Tool Business - The development of humanoid robots poses challenges in design, materials, and manufacturing processes, requiring complex components that can be efficiently produced using five-axis CNC machine tools [5] - The DMU/GMU400 and GMU600 models exhibit unique advantages and competitiveness in processing components for humanoid robots [5] Group 4: Future Strategies in Embodied Intelligence - The company plans to invest in three core technologies: controllers, servo drives, and visual technology, to build a multi-driven system integrating "scenes + data + AI" [6] - The focus will be on enhancing collaboration with AI large model vendors to promote the iterative development and industrial application of intelligent equipment in structured industrial scenarios [6] Group 5: Intelligent Motion Control System - The new generation X5 motion controller has undergone multiple iterations to achieve reliable commercial productization, supporting various mainstream AI computing platforms [7] - The application software platform adopts a cloud-edge-end architecture, providing over 500 underlying functional interfaces and deep integration with mainstream AI large models and simulation systems [8]