How do I connect it to a local llama.cpp instance?
It supports LMStudio or you can start a local endpoint, then run
OPENAI_COMPATIBLE_CUSTOM_KEY="xxx" dirac -y --provider "https://localhost/v1" --model <model_name> "hi..."
alt Hacker News
It supports LMStudio or you can start a local endpoint, then run
OPENAI_COMPATIBLE_CUSTOM_KEY="xxx" dirac -y --provider "https://localhost/v1" --model <model_name> "hi..."