Overview of Ollama
Ollama runs open AI models on your own computer or on its cloud. You start a model from the command line, for example ollama run gemma4, and it also serves a local REST API with Python and JavaScript libraries.
The GitHub README lists installers for macOS, Windows, Linux and Docker, and the project is MIT licensed. It supports model families such as Qwen, DeepSeek, Gemma and gpt-oss, and can launch coding agents like Claude Code and Codex through ollama launch.
Ollama's pricing page adds optional paid plans for cloud models billed per token with monthly usage credits. Developers who want local model serving or private experimentation are the main audience.