Basic LLM Chain node does not support dynamic multi-image input.

Feature Requests
1

The idea is:

The Chat Messages parameter of the Basic LLM Chain node cannot dynamically parse multiple images input by the node.

My use case:

I think it would be beneficial to add this because:

Any resources to support this?

Are you willing to work on this?

2

I’m actually in the same boat with you. I’m performing simple classification based on user’s input and sometimes that contains one or more images. Now I have to do IF but it causes me to duplicate the prompt, the input, etc.

Going forward I’m considering just to do a simple HTTP Request. However that comes at a price of losing all benefits of built in AI nodes.

image
image827×860 38.8 KB