gpt4all

GPT4All is an open source desktop application from Nomic that runs large language models privately on ordinary laptops and desktops. Once the app is installed and a model is downloaded, chat happens entirely on the local machine, with no API calls and no GPU required. The repository describes it as a way to run local LLMs on any device, and it is released under the MIT license, which permits commercial use. It suits individuals, developers and privacy conscious teams who want an offline assistant without sending prompts to a cloud service. Installers are provided for Windows, Windows ARM, macOS and Ubuntu, and a community maintained Flatpak is available. The Windows and Linux builds need an Intel Core i3 2nd generation or AMD Bulldozer class processor or better, and the macOS build needs Monterey 12.6 or newer, with the best results on Apple Silicon. Developers can also install a Python client with pip that wraps llama.cpp implementations and loads models such as Meta Llama 3 variants. The project recently added support for DeepSeek R1 distillations. The software is free. The repository lists no paid plans for the application itself, and the only prices on the page belong to GitHub's own hosting products, not to GPT4All. Note that this is a general local chat tool rather than a dedicated coding assistant, although developers use the Python bindings in their own projects.