Extraction quality <-> Chunk size #551

Open
opened 2026-02-16 00:18:14 -05:00 by yindo · 0 comments
Owner

Originally created by @abelespinc on GitHub (Oct 8, 2025).

If I send a 5-page PDF (product catalog) for extraction, it extracts X products (for example, 50).

If I send those 5 PDF pages one by one for extraction, it extracts Y products (for example, 100).

Why does this happen? Does Llamacloud have a limit? If I send a lot of content, does the extraction get worse?

Originally created by @abelespinc on GitHub (Oct 8, 2025). If I send a 5-page PDF (product catalog) for extraction, it extracts X products (for example, 50). If I send those 5 PDF pages one by one for extraction, it extracts Y products (for example, 100). Why does this happen? Does Llamacloud have a limit? If I send a lot of content, does the extraction get worse?
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: run-llama/llama_cloud_services#551