local-llm-hotkeys

local-llm-hotkeys connects Linux desktop shortcuts to an OpenAI-compatible model server. Its documented desktop setup requires KDE Plasma 6 on Wayland and Python 3.9 or newer. The default backend is FastFlowLM on a compatible AMD Ryzen AI NPU; the README also supports configuring an Ollama server on CPU or GPU. This is desktop software around existing models, not a new model provider.

The MIT-licensed project’s installer links commands into the user’s local bin directory and configures KDE shortcuts and Dolphin menu entries. Automatic dictation pasting can use optional virtual-input permissions; without them, text remains on the clipboard. The security documentation says configured remote servers receive submitted text and screenshots. Local inference depends on keeping the server configuration local.