This project focuses on providing farmers in underserved communities with an intelligent AI-powered chatbot that addresses both general and complex queries related to plant care and farming. The system uses a Strategic agent approach: A local model (LLaMA 3.2 3B) running on an edge device to answer general, everyday plant care questions. A larger, cloud-hosted model (LLaMA 3.1 70 B) accessible via API for handling more complex or specialized queries. An intelligent agent is designed to assess the complexity of each query and assign a score. Based on the score, the agent dynamically routes the query to the appropriate model, optimizing for response time, computational resources, and accuracy. Farmers will interact with the system through a simple web or mobile chat interface (powered by Streamlit), enabling them to ask questions about plant health, harvest cycles, and more.
Category tags: