Text or Image Input

Actually my problem is deeper.
Can't sent any file from Whatsapp (thought Evolution API) to TypeBot.
Is it possible?
Maybe just possible with the official API?

☹️

Can't tell for Evolution as it's not officially supported (you'll have to ask them)

But with the official WA integration, you can definitely collect files. They will be converted as URL

With the official API can I do what I'm trying to?
Let the user and text, image or text + image together?
I want to give the user the option to send images (GPTVision).

This is a problem with evolution. Afaik, the evolution api treatment response coming from type bot is very limited and doesn’t able to handle and format all data from typebot.

I'm not sure if Evolution is not even sending the message to Typebot (cause the unknow format) or if it's sending to Typebot and Typebot is the one returning the error.

Typebot doesn't know how to process image + text so this won't be possible.

How do you expect it to behave in that case?

I'm thinking about add a layer between TypeBot and whatever (Evolution at this moment, but now sure what is the bets option).
So Typebot will always receive a JSON like { tex: 'xxx', image: 'url' }.

Currently the Typebot API only understand some text as input

I'm really not sure how we could support hybrid messages 🤔

It also supports files, as we already have the File Upload block. However, it currently lacks a component that allows users to input text, upload a file, or combine text and file inputs—a feature similar to the interface I'm currently using here to type this message.

With the rising popularity of GPTVision, coupled with the fact that it is becoming more cost-effective, I believe that all agents will inevitably transition in this direction.