The official Meta cookbook of scripts and notebooks for fine-tuning, evaluating and deploying Llama models.
| Category | Fine-tuning |
| Type | Recipes & scripts |
| License | MIT |
| Runs locally | Yes |
| Built with | Python |
| Skill level | Intermediate |
| Best for | fine-tuning Llama models the supported way |
Other open-source fine-tuning tools worth comparing:
UnslothFine-tune LLMs 2x faster on one GPU
AxolotlConfig-driven fine-tuning for many models
LLaMA-FactoryFine-tune 100+ models with a UI
PEFTLoRA and friends from Hugging Face
TRLAlign LLMs (SFT, DPO, PPO)
TorchtunePyTorch-native post-training, hackable recipes
ms-swiftFine-tune 500+ LLMs and VLMsLlama Cookbook is free and open-source (MIT license), so you can use, self-host and modify it at no cost.
Yes. Llama Cookbook is designed to run on your own machine or server, keeping your data private.
Popular open-source alternatives include Unsloth, Axolotl, LLaMA-Factory. See the comparisons above to choose.
Browse the full directory of open-source AI tools, models and projects — updated daily.
Browse all tools →