Find answers from the community

Updated 2 months ago

Good practices for chatting with pdf slides using llamaindex

At a glance

The community member is looking for a tutorial on how to chat with PDF slides using LlamaIndex, including slides with tables and infographics. The comments suggest trying LlamaParse, which supports ingesting .ppt and .pptx files, but the community member notes they have PDF slides created with Beamer/LaTeX and will investigate further. There is no explicitly marked answer in the comments.

Useful resources

aadenml

Is there any tutorial on good practices on how to chat with pdf with slides using llamaindex? The slides might have tables, infographics etc? How should I ingest them, with what models, what tools?

6 comments

WWhiteFang_Jr

You should try LlamaParse: https://docs.llamaindex.ai/en/stable/llama_cloud/llama_parse/

aadenml

I didn't see anything related to slides

WWhiteFang_Jr

we do support slides:
.ppt: https://github.com/run-llama/llama_parse/blob/4897d01cb075ed0835b7df0d072a7ad4e39c4e64/llama_parse/utils.py#L122
.pptx: https://github.com/run-llama/llama_parse/blob/4897d01cb075ed0835b7df0d072a7ad4e39c4e64/llama_parse/utils.py#L124

aadenml

Oh thx, looks cool

aadenml

Although I have pdf with slides (like those created with beamer / latex)

aadenml

But I will investigate further

Add a reply