Get started with WidePane
Install the app, connect a model, and you're coding with AI in minutes — locally or with your own cloud keys.
Install
Get the Windows desktop app, or the free VS Code extension (no account). Download the
MSI or NSIS installer from Downloads, or run
ext install widepane.widepane in VS Code.
Connect Ollama (local)
Run models entirely on your machine. Install Ollama and pull a model, e.g.
ollama pull mistral. In WidePane, set the Ollama base URL (default
http://127.0.0.1:11434). To use an Ollama server elsewhere on your
network, enter its address — WidePane classifies it as “On your network” and requires
explicit approval in local‑only mode.
Bring your own cloud key (optional)
OpenAI, Anthropic, or any OpenAI‑compatible endpoint. Add your API key in Provider settings. It's stored in your OS keychain and used only by the trusted host — never exposed to the UI. Choose which model handles each role (agent, edit, chat, completion).
Work
Open a folder, ask questions in the sidebar, use inline edit on a selection, or run Agent mode to make multi‑step changes — every action is reviewed before it applies.
Reference docs
In‑repo documentation covers the deeper details.