It is based on large language model internship to read the image data The packages used it are google.generativeai, Streamlit and PIL First you need to install !pip install google.generativeai And then import the packages The Model used is "model=genai.GenerativeModel("gemini-1.0-pro-vision-latest")" Gemini - pro - Vision - Latest model we handles image data API key you need get it from google.aistudio where we need click on get-api key