Skip to content

Explore Llama Parse #255

@ppinchuk

Description

@ppinchuk

We should consider having Llama Parse as a PDF and doc read option, especially since the table reading accuracy could be significantly improved with it.

Since it's expensive, we need to be strategic about where to apply it. At a minimum only for PDF pages that pass the heuristic, but maybe even better if we restrict it to pages with tables or something. Explore this trade-off

Metadata

Metadata

Assignees

Labels

enhancementUpdate to logic or general code improvementsnew computationUpdate that adds a new computation methodp-highPriority: hightopic-python-generalIssues/pull requests related to python

Projects

No projects

Relationships

None yet

Development

No branches or pull requests

Issue actions