https://github.com/lexiestleszek/sova_ollama

Open source implementation of Sova - RAG-based Web search engine using power of LLMs. Using Langchain, Ollama, HuggingFace Embeddings and scraping google search results.
https://github.com/lexiestleszek/sova_ollama

large-language-models llm rag-implementation retrieval-augmented-generation web-scraping

Last synced: about 2 months ago
JSON representation

Open source implementation of Sova - RAG-based Web search engine using power of LLMs. Using Langchain, Ollama, HuggingFace Embeddings and scraping google search results.

• Host: GitHub
• URL: https://github.com/lexiestleszek/sova_ollama
• Owner: LexiestLeszek
• License: mit
• Created: 2024-02-10T12:19:52.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-02-12T10:29:27.000Z (11 months ago)
• Last Synced: 2024-04-28T05:06:10.134Z (9 months ago)
• Topics: large-language-models, llm, rag-implementation, retrieval-augmented-generation, web-scraping
• Language: HTML
• Homepage: https://sova.fyi
• Size: 32.2 KB
• Stars: 7
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
### Sova

## Description

Sova is a web search engine that utilizes power of Large Language Models (LLM) to return full text answer instead of bunch of links like traditional search engines do.

## Installation

Before installing Sova, ensure you have Python installed on your system. You will also need pip, which is typically included with Python.

To install Sova, follow these steps:

1. Clone the repository using this command in terminal: `git clone https://github.com/LexiestLeszek/sova_ollama.git`

2. Install the required dependencies using this command in terminal: `pip install -r requirements.txt`

3. Install Ollama from their website https://ollama.com/

4. Download the local LLM (in our case - qwen:1.8b) using this command in terminal: `ollama pull qwen:1.8b`

5. Run the gunicorn server using this command in terminal: `gunicorn -b 0.0.0.0:8080 app:app`

6. Use the search-bar to input your search query or or question.