Hi there folks! 👋 Does anyone know

There are a few different options:

https://docs.llamaindex.ai/en/stable/examples/response_synthesizers/tree_summarize.html

https://docs.llamaindex.ai/en/stable/api_reference/indices/list.html

For multimodal:

https://docs.llamaindex.ai/en/stable/examples/multi_modal/openai_multi_modal.html

this is awesome! let me read all of this!

I think my flow would be something like converting pdfs to markdowns, while extracting the images to a markdown relative path, then run the multimodal to replace the images with their descriptions, and finally run a summarization of the markdown to a condensed one.

Am I on the right path here?

Yeah this part covers the usage of the PDFs if that's where your images are mostly contained: https://docs.llamaindex.ai/en/stable/examples/multi_modal/multi_modal_pdf_tables.html

Maybe you could just directly use that, at least worth a try