ollama/ollama
Ollama is an open-source platform designed to run and serve large language models locally or in cloud environments. The README uses Gemma as an example and links to the official model catalog. Ollama provides a command-line interface, a local REST API endpoint on port 11434, official client libraries for Python and JavaScript, and integrations with developer tools like Claude Code, Codex, Copilot CLI, and OpenClaw. The backend relies on the llama.cpp project. The GitHub snapshot shows the repository is not archived.
Check platform requirements in the official documentation before installing. On macOS and Linux, installation executes via curl -fsSL https://ollama.com/install.sh | sh, alongside standalone installer downloads. On Windows, setup runs through PowerShell using irm https://ollama.com/install.ps1 | iex or an executable installer. Container deployment uses the official Docker image ollama/ollama. Programmatic integrations install via pip install ollama or npm i ollama. The project is distributed under the MIT license. For comparison, llama.cpp provides LLM inference tools; Ollama also lists it as a backend, with a different workflow.
Data as of : 185743 stars · 18885 forks.