Llama Cookbook
A collection of notebooks and guides for running, fine-tuning, and building apps with Meta's Llama models.
- Stars
- 18.6k
- Forks
- 2.8k
- Licence
- MIT
- Language
- Jupyter Notebook
- Last update
- 6 months ago
- Type
- Learning resource
What it is
This is a repository of example notebooks and guides from Meta showing how to work with the Llama family of AI models. It covers getting started with inference (running a model), fine-tuning (adapting a model to your data), retrieval-augmented generation (RAG), and complete end-to-end sample projects. The README says it also includes recipes for using Llama models through various provider services.
- Follow a notebook to run a Llama model for inference
- Work through a fine-tuning example to adapt a model
- Explore an end-to-end example such as the research paper analyzer or book character mind map
- Try the Llama API getting-started notebook
Great for
- Developers who want working code examples for running Llama models
- People learning how to fine-tune a language model on their own data
- Builders interested in Llama 4 examples such as the Llama API and WhatsApp bot integration
- Engineers comparing how Llama is used through different hosting providers
Not ideal for
- Non-programmers, since the material is Jupyter notebooks and Python code
- People looking for a finished app to use directly, because this is a set of examples rather than a product
- Anyone needing a stable library API, since the README notes a recent repo refactor and the latest release is v0.0.5
How to get it
- 1
Clone and build from source
You'll need: Python · Jupyter Notebook
Clone the repository
git clone https://github.com/meta-llama/llama-cookbook - 2
Run it
- Clone the repository from GitHub
- Open the getting-started folder and pick a notebook
- Read the notebook's setup notes for required packages and accounts
- Run the notebook cells in Jupyter
- 3
Get the most out of it
- Start with the Llama API getting-started notebook to learn the basics
- Read the Fine-Tuning FAQ in the src/docs folder for common questions
- Check the 3P Integrations folder for provider-specific recipes
- Review the model cards and license links for the Llama version you use
Downloads
Strengths
- Covers inference, fine-tuning, and RAG in one place
- Includes examples for the newest Llama 4 models according to the README
- Released under the MIT license
Trade-offs
- The README says the repo was recently refactored and some links or folders may have moved
- Using the Llama models themselves is governed by separate Meta license and acceptable use policies
- Examples target specific provider services and may need accounts or API keys the README does not detail
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| llama-cookbook (this) | 18.6k | Learning resource | intermediate | 6 months ago | Developers who want working code examples for running Llama models |
| annotated_deep_learning_paper_implementations | 67.6k | Learning resource | intermediate | 4 days ago | Students learning how transformers, GANs, or optimizers work internally |
| Made-With-ML | 49.9k | Learning resource | intermediate | 8 months ago | Software engineers who want to learn how ML models get deployed and maintained |
| pytorch-deep-learning | 19.3k | Learning resource | beginner | 1 months ago | Beginners in machine learning who want to learn PyTorch in a hands-on, code-first way |
| FinGPT | 21.5k | AI model / agent | advanced | 12 days ago | Developers who want to experiment with finance-tuned LLMs for sentiment analysis |
| LaTeX-OCR | 16.6k | AI model / agent | intermediate | 1.8 years ago | Students who want to turn a handwritten or printed formula photo into editable LaTeX |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.