This application is a CLI chatbot for SDK/Hardware documents that uses the Local RAG model with LLama3 with Ollama, LanceDB, Openhermes Embeddings.
The chatbot is built using the RAG mode using Phidata Assistant and Knowledge Base.
- Install Dependencies
pip install -r requirements.txt
- Setup Ollama
- Install Ollama on Linux
curl -fsSL https://ollama.com/install.sh | sh
- Install Ollama on Mac
brew install ollama
- Run Llama3 and Openhermes with Ollama
ollama run llama3
ollama run openhermes
Now you are ready to run CLI SDK manual Chatbot locally
python3 assistant.py
Note: This Chatbot utilizes data/manual.pdf
. For running it on your document change path of document in assistant.py