Llama Stack Tutorial

Welcome to the Llama Stack Tutorial - your comprehensive guide to learning and using Llama Stack.

What is Llama Stack?

Llama Stack is an open-source framework that makes it easier to build, run, and experiment with large language models on your own infrastructure. It provides:

• Unified interface for different model providers (Ollama, vLLM, etc.)
• Tools and agents for complex AI workflows
• Safety and telemetry built-in
• CLI, Python SDK, and web playground for development

Tutorial Content

This repository contains hands-on tutorials covering:

🟢 Beginner

• Getting started with CLI and Python SDK
• Using the web playground
• Basic model interactions

🟡 Elementary

• Model Context Protocol (MCP) integration
• Retrieval Augmented Generation (RAG)
• Building interactive agents with ReAct

🟠 Intermediate

• Safety and content filtering with Llama Guard
• Telemetry and observability with OpenTelemetry

🔴 Advanced

• Model evaluation with built-in framework
• All-in-one deployment setup

Access Points

• Web Playground: http://localhost:8502
• Llama Stack API: http://localhost:8321
• Jaeger Telemetry: http://localhost:16686
• Tutorial Site: Run ./utilities/lab-serve then http://localhost:8080

Contributing

This tutorial is designed to help developers learn Llama Stack through practical examples. The content is built with Antora and can be edited in the content/modules/ROOT/pages/ directory.

---