Web Query System with ChatGPT4o-mini with OpenaAI API

Web Query System based on ChatGPT4o-mini, accessible through a Telegram chatbot. The idea is to harness the power and cheapness of the ChatGPT4o-mini language model to deliver precise, up-to-date and useful information directly to you with minimum hallucinations, mainly because information is scraped from the web.

How It Works

The process involves several steps to ensure that you get the most relevant information:

1. Optimization: The query is optimized using the aforementioned LLM model for effective Google searches.
2. Search Execution: A request is made to the Google Search API to retrieve data.
3. Data Extraction: The top three search results are scraped and the content is saved in a .txt file.
4. Data Processing: The content from the .txt file is embedded with the OpenAI Embeddings API, vectorized, and queried against the 4o-mini model to extract appropriate information and answer the query.

Deployment and Technology

• Hosting: The system is deployed on a Google VM instance using a Docker image.
• API Integration: The 4o-mini model is integrated through the OpenaAI API.

Check out the demo here - https://youtu.be/NKYEsZQmUME