Connection Setup

  1. Download the KoboldCPP application.
  2. Launch KoboldCPP to host the GGML/GGUF model.
  3. Configure the connection to the KoboldCPP endpoint in Settings (modifiable at any time to switch models).