Awesome

Ollama RAG

Project for adding RAG to ollama models using :

• langchaingo
• qdrant vector database
• nomic-embed-text
• Mistral

Installation :

Download packages :

go mod download

Pull Ollama models :

ollama pull nomic-embed-text
ollama pull mistral

Install qdrant :

docker pull qdrant/qdrant
docker run -p 6333:6333 qdrant/qdrant

Create Collection for qdrant use any http client for make a PUT request like example blow for creating a Collection

curl -X PUT http://localhost:6333/collections/romeo \
  -H 'Content-Type: application/json' \
  --data-raw '{
    "vectors": {
      "size": 768,
      "distance": "Dot"
    }
  }'

Delete a collection

curl --location --request DELETE 'http://localhost:6333/collections/romeo'

Using

put you texts in text.txt and run : go run . then ask anything from the text you provided

TODOS:

• adding http request to make collection with functions
• dynamic file selection
• check if some file is embedded and already exist in vector database