Ollama
Runs open-weight models locally behind a simple command and HTTP API.
Überblick
Ollama downloads, quantises and serves open-weight models on a local machine with one command, exposing an HTTP API compatible with the hosted providers. It solves the setup problem: running a model locally otherwise means choosing a runtime, converting weights and managing memory yourself. It exists because that friction is what stops people trying local inference at all, and it is the usual first step before moving to a more configurable server.
Technologiegraph
Wie Ollama mit dem übrigen Stack zusammenhängt.
Werkzeuge
Software, die Ollama implementiert oder integriert.