# LlamaFarm

An open-source AI platform that runs entirely on local hardware for building RAG applications, training custom classifiers, detecting anomalies and processing documents, with no cloud dependency.

LlamaFarm is an AI platform that runs on a user's own hardware rather than a cloud service, aimed at retrieval-augmented generation over PDFs, documents and CSVs, training custom text classifiers from as few as 8 to 16 examples using SetFit, batch and streaming anomaly detection with more than a dozen algorithms, OCR and structured-data extraction from images and PDFs, named entity recognition, and tool calling through MCP.

It ships as a desktop app for Mac, Windows and Linux with no command line required, or as a CLI that scaffolds a project with lf init, starts local services and a Designer web interface with lf start, and opens an interactive or one-off chat with lf chat. It can switch between Ollama, OpenAI, vLLM or local GGUF models as its runtime, with automatic GPU and NPU acceleration on Apple Silicon, NVIDIA and AMD hardware.

Website: https://llamafarm.dev