Planes, subways, mountains — most of Kition runs offline. The only thing that needs network is your AI provider.
Configure a local Ollama / vLLM provider. Model quality is a notch below cloud, but fully offline at zero privacy cost.
Models & providers
Under China network conditions, domestic models are often the most stable choice. Setup mirrors OpenAI — one baseURL away.
Getting started
One table on wire types, real configs for three providers, and a checklist for auth failures.
Kition is a local-first AI workspace. Markdown documents, structured tables, and an AI agent — running on your own machine, against the model provider you choose.
