Ollama

Ollama is a free, open-source tool for running open large language models directly on your own machine. Instead of sending prompts to someone else's server, you download a model once and run it locally, on your own hardware. It installs on macOS, Windows, and Linux from ollama.com.

The appeal is control and privacy. Once a model is downloaded it runs offline, and your prompts never leave your computer — which matters for sensitive or confidential work. A single command (for example, 'ollama run' followed by a model name) pulls a model and starts a chat; Ollama also exposes a local API so developers can build applications against locally-hosted models, and it works with many third-party front-ends.

It fits developers, privacy-conscious users, and anyone who wants to experiment with open models without ongoing per-use costs or a network dependency — people comfortable installing a tool and, ideally, with a reasonably capable computer.

To begin, download Ollama from ollama.com, install it, and run a model from the terminal; it is genuinely free, with no account or credit card required. Starting with a smaller model is the sensible first step.

The honest limits, stated plainly. Ollama is free, but you supply the computer — large models need substantial memory and benefit greatly from a capable GPU, and a modest machine is limited to smaller models that are less capable than the big hosted assistants. It is more technical than a web chatbot, and the quality and licensing of each open model vary, so check a model's licence and card before relying on it.