# dumbpilot

Get inline completions using llama.cpp as as server backend

## Usage

1. start llama.cpp/server in the background or on a remote machine
2. configure the host
3. press `ctrl+shift+l` to use code prediction