/gpt-researcher-ollama

Based on assafelovic/gpt-researcher - Modified to support local Ollama models

Primary LanguagePythonMIT LicenseMIT

🔎 GPT Researcher

This is a fork from https://github.com/assafelovic/gpt-researcher with some modifications to make it work with local Ollama models.

How to use

  1. Clone this repository
  2. Install the requirements
  3. Copy the .env.example file to .env and fill in the fields accordingly
    • Make sure to have the Ollama model already downloaded with ollama server running
    • The example is using mixtral:8x7b-instruct-v0.1-q6_K
  4. For embbedings I'm using llama3, so unless you change gpt_researcher/memory/embeddings.py to use something else you will need to run: 
    ollama pull llama3
  5. Modify the query variable in the main.py file with the prompt you want to use for your research
  6. Run the main.py file