SimpleRAG

This is a simple RAG sample based on:

• LangChain
• Ollama
• Chroma

Which can be run by using:

• Streamlit
• React + FastAPI

Install Ollama

Deploy models

Install python libraries

pip install -r requirements.txt

Config the app

Edit .env file in the src directory.

VECTORDB_PERSIST_DIRCTORY = "../vectordb"
LLM_MODEL_NAME = "deepseek-r1:1.5b"

Run Streamlit

cd src

streamlit run app.py

Run FastAPI

uvicorn api.app:app

Acess the API docs page at http://localhost:8000/docs

Run React

cd frontend

npm start

Access the web page at http://localhost:3000