35 lines
1.1 KiB
Markdown
35 lines
1.1 KiB
Markdown
# RAG Agent with Local Qdrant and Ollama
|
||
|
||
This repository implements a simple AI agent that can store, search, and retrieve information from a local vector store using Qdrant and Ollama embeddings. The agent is built with LangChain v1 and supports an interactive CLI with the following commands:
|
||
|
||
* `/add` – add a new document to the knowledge base.
|
||
* `/search` – perform a semantic search in the knowledge base.
|
||
* `/quit` – exit the program.
|
||
|
||
## Features
|
||
|
||
* **RAG** – Retrieval-Augmented Generation using a local vector store.
|
||
* **Qdrant** – Vector similarity search engine.
|
||
* **Ollama** – Local LLM (`llama3`) and embeddings (`nomic-embed-text`).
|
||
* **LangChain v1** – Modern agent framework.
|
||
* **Recursive text splitter** – Chunk documents before embedding.
|
||
|
||
## Setup
|
||
|
||
```bash
|
||
# Install Ollama models
|
||
ollama pull llama3
|
||
ollama pull nomic-embed-text
|
||
|
||
# Install Python dependencies
|
||
pip install -r requirements.txt
|
||
```
|
||
|
||
## Usage
|
||
|
||
```bash
|
||
# Load documents from the `docs` folder and start the CLI
|
||
python -m src.cli --docs docs
|
||
```
|
||
|
||
You can then interact with the agent using the commands described above. |