Custom RAG

RAG using ollama on local machine with Llama 3.2 1B model and contents from web pages scraped on the fly.

Using this repo

• Clone the repo

git clone https://github.com/turinaf/rag_ollama.git

Just add links of website from where you want to get the knowledge base to flask_api/embed_data.py and it will load page content from the link using langchain's WebBaseLoader .

• Then start docker app and build docker container to run it

docker compose up --build

After building successfully visit http://localhost:8080/ to chat with your custom RAG