I am currently in the process of developing the chatbot that utilizes your personal data to answer contextually. I am already halfway through the project, and I plan to open-source the code soon.
- Node.js, Python, MongoDB, and React for the web application
- OpenAI ChatGPT Turbo 3.5 for conversation
- OpenAI Ada model for embedding
- Vectorization for long-term and short-term memory
- Pincones Database for embedding
- User management system
- Long-term, permanent conversations with the bot
- Custom indexes - use your own files, PDFs, TXT files, websites
- Automatic retry on API errors
- Change and view model parameters such as temp, top_p, etc. directly