Looks really nice! How does it handle tables?
We have two ingestion pathways: 1. regular OCR + text embeddings; 2. Colpali. We've observed that Colpali does a much better job with tables since it can encode positional stuff and layouts as well.
alt Hacker News
We have two ingestion pathways: 1. regular OCR + text embeddings; 2. Colpali. We've observed that Colpali does a much better job with tables since it can encode positional stuff and layouts as well.