aatmaj28/PageIndex-Comparative-Study

Benchmarking PageIndex (vectorless tree-based RAG), DIY vectorless RAG, and traditional vector RAG (Qdrant) on a 123-page SEC 10-K filing. 10 retrieval questions testing tables, footnotes, and cross-section reasoning. Tree-based approaches scored 100% vs vector RAG's 60%.

14
/ 100
Experimental
No License No Package No Dependents
Maintenance 13 / 25
Adoption 0 / 25
Maturity 1 / 25
Community 0 / 25

How are scores calculated?

Stars

Forks

Language

Python

License

Last pushed

Apr 07, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/aatmaj28/PageIndex-Comparative-Study"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.