{"repo":"hrishi-008/SummarAI","free":true,"listed":false,"github":"https://github.com/hrishi-008/SummarAI","clone":"git clone https://github.com/hrishi-008/SummarAI.git","description":"A tool for summarizing search results and website content using FAISS, LLMs, and the Retrieval-Augmented Generation (RAG) technique.","language":"Python","stars":30,"topics":["vector-database","faiss","faiss-vector-database","groq","llm","rag","summarization"],"license":null,"category":"ai-agents","readme_excerpt":"SummarAI (earlier SearchGPT) A tool for summarizing search results and website content using FAISS, LLMs, and the Retrieval-Augmented Generation (RAG) technique. Overview Search GPT is a Streamlit-based application that automates the process of scraping Google search results, extracting content from websites, storing and retrieving information using FAISS (Facebook AI Similarity Search), and summarizing the content using a language model. This tool leverages the Retrieval-Augmented Generation (RAG) technique to provide concise and easy-to-understand summaries of web content based on user queries. While the github version uses FAISS, the dockerised veriosn uses ANNOY (Approximate Nearest Neighbors Oh Yeah) for retrieval, everythig else remains same. This is done to improve the speed of making the vector DB, althuogh at the cost of the quality of retrieved documents. Demo Search GPT Demo Features - Google Search Scraping : Automatically scrapes Google search results for a given query. - Website Content Extraction : Scrapes and extracts relevant content from the search result links. - FAISS Database Integration : Stores and retrieves scraped data efficiently using FAISS. - Content Summarization with RAG : Uses the RAG technique to generate summaries by combining retrieved content with a language model, ensuring accurate and context-aware summaries. - Streamlit Interface : Provides a simple web interface for users to input queries and receive summarized results. Dockerization The","default_branch":null,"files":null,"tree":[],"storefront":"/r/hrishi-008","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/hrishi-008/SummarAI/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}