Closed drusepth closed 1 year ago
I think we want to wait for local in-browser LLM support (a la WebGPU, being enabled in Chrome on the 24th) for this feature so we don't dump users into an app that requires a paid OpenAI key also. Otherwise, we might end up just hosting our own LLM API that Premium users get access to and don't need to manage a separate API key for.
WIP for #1245