TLDR: China Agricultural University has launched its third-generation Shennong Large Model, an advanced AI system designed to significantly enhance farming practices and agricultural research. This model, introduced at the 2025 World AgriFood Innovation Conference, is China’s most comprehensive agricultural AI, featuring 36 intelligent agents for efficient and accessible farm management and pest control.
Beijing, China – October 15, 2025 – China Agricultural University (CAU) today announced the official launch of its groundbreaking Shennong Large Model 3.0, marking a pivotal moment in the intelligent transformation of global agriculture. Unveiled at the prestigious 2025 World AgriFood Innovation Conference in Beijing, the third-generation Shennong model is touted as China’s most comprehensive and data-rich artificial intelligence system dedicated to the agricultural sector.
According to Sun Qixin, president of China Agricultural University, the Shennong Large Model 3.0 is the nation’s first large-scale AI system trained on the country’s most extensive agricultural dataset to date. This robust dataset encompasses an impressive 90 percent of agricultural disciplines and covers 80 percent of agricultural scenarios, ensuring a broad and deep understanding of diverse farming challenges.
The new 3.0 version is equipped with 36 intelligent agents, emphasizing a design philosophy focused on ‘small size, high intelligence, and low cost.’ This approach aims to make agricultural AI more practical and widely accessible to a vast network of farmers and production entities. Wang Yaojun, an associate professor at CAU’s College of Information and Electrical Engineering and leader of the development team, highlighted the model’s user-friendly nature. ‘Through a mobile app or mini program, farmers can access AI-powered guidance for crop management and pest control within seconds,’ Wang stated.
One of the standout features is the pest and disease identification agent, capable of recognizing 70 categories and over 600 types of crop pests and diseases. This agent provides instant, expert-level advice, comparable to consulting multiple plant protection specialists. Beyond immediate farm-level applications, President Sun Qixin noted that the model is also designed to accelerate scientific discovery for agricultural researchers and support data-driven breeding decisions.
The Wheat Breeding AI Assistant, one of the 36 intelligent agents, exemplifies this advanced capability by integrating national-level germplasm data and expert knowledge. This specialized assistant aims to shift wheat breeding from traditional trial-and-error methods to innovative, data-based approaches. The Shennong Large Model series has seen rapid evolution since its initial version in 2023, with this latest iteration named after Shennong, a legendary figure in ancient China revered for his contributions to farming.
Also Read:
- Generative AI Propels Sophisticated Academic Fraud in Chinese Paper Mills
- Indonesia Advances Agentic AI Integration for Enhanced Policy Analysis and Public Sector Efficiency
The 2025 World AgriFood Innovation Conference, which hosted nearly 800 participants from dozens of countries, focused on critical discussions surrounding resilient food supply and the transformation of agrifood systems, providing an ideal platform for the unveiling of this significant AI advancement.
