Wanlian Securities: Manus Expected to Become the Paradigm of Next-Generation Human-Machine Collaboration, Paying Attention to Accelerating the Landing of AI Applications

Wanlian Securities released a research report stating that the upgrade of open source models such as DeepSeek R1 and V3 is expected to accelerate the landing of industry-specific AI models and AI application products represented by AI Agents. Recently, Chinese AI startup company Monica released the world's first general AI Agent product, Manus, which has five major advantages such as autonomous execution ability, continuous learning, and memory. Manus surpasses OpenAI Deep Research in performance testing and is expected to become the paradigm of the next generation of human-machine collaboration. Recommendations include focusing on the landing of industry-specific AI models, commercialization of AI application products such as AI Agents, application of AI models and AI Agents on the edge, and the boost in computing power demand due to the wide-ranging application scenarios of AI Agent products represented by Manus. Key points of Wanlian Securities: Chinese startup company Monica releases the world's first general AI Agent Manus with 5 main advantages and characteristics. Chinese AI startup company Monica has released its product Manus, according to official introductions, Manus is a general AI Agent that can connect thoughts and actions. It not only thinks but also provides results. Compared to other AI products, Manus' main advantages and characteristics include: 1) Autonomous execution ability: Manus can independently complete tasks in the cloud without human intervention and deliver complete task results directly; 2) Multi-agent systems: Supported by a variety of models, it has powerful tool invocation capabilities, can flexibly write code, intelligently browse the web, and operate various applications; 3) Continuous learning and memory: Manus has memory function and can optimize task execution strategies based on historical data and experience; 4) Wide range of application scenarios: including research, life, data analysis, education, productivity tools, and creative entertainment, with a total of 51 specific use cases; 5) Efficient human-machine collaboration: creating a new paradigm of human-machine collaboration, users only need to describe their needs in natural language, and Manus can quickly respond and execute tasks. Manus surpasses OpenAI Deep Research in performance testing and is expected to become the paradigm of the next generation of human-machine collaboration. AI Agent is currently one of the mainstream trends in AI large model products, and autonomy is its main feature. The advancement of AI Agents represents a more autonomous approach compared to previous AI applications such as chatbots and AI search products. From the achievements of intelligent spectrum AI in the Agent stage to the case studies of Manus in office scenarios, human resources, financial analysis, and other fields, the characteristics of autonomy and wide application scenarios of AI Agents can be demonstrated. This performance breakthrough in the AI Agent Manus test, surpassing OpenAI Deep Research in all three difficulty levels, is seen as a step towards the next generation of human-machine collaboration paradigm. Risk factors: Sino-US technology friction, competitiveness of domestic AI large models lower than expected, AI Agent application landing lower than expected, and AI industry technological development and computing power demand lower than expected.