llama-cookbook
Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started with Inference, Fine-Tuning, RAG. We also show you how to solve end to end problems using Llama model family and using them on various provider services
Official Meta collection of guides and code examples for building applications with Llama AI models, covering inference, fine-tuning, and retrieval-augmented generation.
The Llama Cookbook is the official collection of guides and example code maintained by Meta for building applications using the Llama family of AI language models. It is primarily organized as a set of Jupyter notebooks (interactive documents that combine code and explanations) and covers three main areas: running inference (getting a model to generate responses), fine-tuning (adapting a pre-trained model to a specific task with new data), and retrieval-augmented generation (a technique that lets a model answer questions by first searching a document library for relevant information).
Beyond those fundamentals, the repository contains examples of complete end-to-end applications — such as building a chatbot integrated with WhatsApp, analyzing research papers, or generating character relationship maps from a novel. A separate section covers integrations with third-party hosting providers and services.
The repository was previously called llama-recipes and was renamed to its current name. It supports multiple generations of the Llama model family. Each model version has its own license that must be reviewed separately before use.
Where it fits
- Build a WhatsApp chatbot that answers customer questions using Llama.
- Fine-tune Llama on your company's documents to answer domain-specific questions.
- Create a research paper analyzer that summarizes and extracts insights from academic PDFs.
- Generate character relationship maps and story summaries from novels using retrieval-augmented generation.