Install Backends Tools Config

rag

Agentic RAG Transformer for intelligent knowledge retrieval

Semantic search powered by FAISS and sentence transformers. Ask questions, run calculations, search the web, and more.

Get Started GitHub

Installation

Set up rag in seconds

rag works out of the box with minimal configuration. Choose quick setup or manual installation.

Quick Install
bash SETUP.sh
Manual
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
pip install -e .

Run the interactive minimal TUI:

$ rag-agent

Or with rag-tui:

$ rag-tui --theme minimal

LLM Backends

Choose your preferred model

rag supports multiple backends. OpenAI for best quality, Cerebras for speed, Ollama for offline use.

OpenAI

GPT

Cerebras

API

Ollama

Local

Local

HF

Tool Commands

Built-in utilities

Extend rag with built-in commands. Run calculations, search Wikipedia, execute shell commands, and more.

CALC:
Calc
TIME:
Time
WIKI:
Wiki
SHELL:
Shell
SEARCH:
Search
WEB:
Web

Configuration

Customize rag behavior

Configure rag via environment variables. Set the LLM backend, embedding model, memory mode, and API keys.

BACKENDlocal
MEMORYsession
MODELMiniLM-L6
OPENAIsk-...
CEREBRAScs-...
TMDB...
TOP_K3
MAX_LEN150
FAISS1