This looks neat. Thanks for providing an open source alternative!
However, you say everything is "100% local", but then require an API key for one of the big 4 LLM providers. Anything leaving my PC isn't local. Could you add the option to use a local model (ie. Ollama or custom API URL) for inference? That would truly make it local first, and a great companion app.
However, you say everything is "100% local", but then require an API key for one of the big 4 LLM providers. Anything leaving my PC isn't local. Could you add the option to use a local model (ie. Ollama or custom API URL) for inference? That would truly make it local first, and a great companion app.