Free, open-source apps and engines for running large language models on your own computer or server. Offline, private, and no API bill.
A simple CLI and local server for downloading and running open-weight models.
An open-source desktop chat app that runs fully offline.
Runs local LLMs on everyday hardware. Licensed for commercial use.
A local-first app for chatting with your documents, with agents built in.
A self-hosted web chat interface that works with Ollama and OpenAI-compatible APIs.
A popular local LLM web UI with an OpenAI-compatible API.
Runs GGUF models with the KoboldAI UI. One file, nothing to install.
A chat and roleplay frontend for power users. Connects to local or remote backends.
LLM inference in C/C++. The engine behind most GGUF tooling.
Bundles a model and its runtime into a single executable file.
A drop-in local API for text, vision, voice and image models. No GPU required.
A high-throughput, memory-efficient serving engine for LLMs.
A fast serving framework for language and multimodal models.
Compiles and deploys LLMs across GPUs, phones and browsers.
Runs and fine-tunes LLMs on Apple silicon with MLX.
Splits large models across your own devices so you can run them at home.
Every link goes to the project's own GitHub repo or official site. Check each project's license and docs before you use it.