How to deal with multimodality in TextGrad

[felixgao]

Hi,

I have many different documents that is in JPG, PNG, or PDF format that I need to extract certain information from them. How do I use TextGrad to optimize the LLMs with multimodality?

Thanks

[mertyg]

Currently, TextGrad only supports text.
However, we’ll have news on this soon! :)

[mertyg]

It's here now! We'll soon push some benchmark results too. :)