The idea is:
Just allow connection of the code tool to the embed and model inputs of AI nodes
My use case:
This would allow a lot more functionality. We could call any api endpoint. So using things like together.ai, or other local models besides Ollama. Would also allow for processing of results like reranking.
I think it would be beneficial to add this because:
It would satisfy basically all requests for using different services and endpoints. Allowing the dev team to focus on adding nodes only for the most popular services.
Any resources to support this?
Are you willing to work on this?
Near as I can tell all it would require is letting the already existing code tool connect to the ai node ports, and giving some documentation on what that data format needs to be.