Thankyou for the great suggestion joelthelion! The component currently relies on calling out to external services to interact with their models, but I am planning to add functionality to host LLAMA models entirely on the browser. I don't think any of them have RAG capabilities but if you have suggestions how this could be achieved that would be greatly appreciated! Thanks again joelthelion!
4 comments
[ 3.0 ms ] story [ 19.9 ms ] thread