VectifyAI/PageIndex

đź“‘ PageIndex: Document Index for Vectorless, Reasoning-based RAG

65
/ 100
Established

Builds a hierarchical tree-index from documents—similar to a machine-generated table of contents—then uses LLM reasoning to traverse the tree for retrieval, eliminating the need for vector databases or artificial chunking. Achieves 98.7% accuracy on FinanceBench by reasoning over document structure rather than semantic similarity. Integrates via self-hosted Python, MCP protocol, or cloud API, with support for vision-based retrieval directly from PDF page images.

21,374 stars. Actively maintained with 21 commits in the last 30 days.

No Package No Dependents
Maintenance 20 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

21,374

Forks

1,665

Language

Python

License

MIT

Last pushed

Mar 04, 2026

Commits (30d)

21

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/VectifyAI/PageIndex"

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