VectifyAI/PageIndex

Health v2 · maintenance only

Adoption not part of the score

38442 stars · 3331 forks observed · 2026-10-02

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-28, confidence 0.97

PageIndex is a Python SDK that provides a vectorless, reasoning‑based retrieval engine for long documents. It builds a hierarchical tree index and lets an LLM reason over it to return context‑aware answers.

Use cases

• retrieve relevant sections from lengthy PDFs without using a vector database
• build a chat assistant that can answer questions over complex corporate reports
• enable context‑aware document search in enterprise knowledge bases
• replace traditional chunk‑and‑embed pipelines with tree‑based indexing for higher relevance
• run local, privacy‑preserving RAG on confidential documents

When to choose

• you need explainable, traceable retrieval that mimics human reading
• you want to avoid the overhead of maintaining a vector database
• your documents are long, structured, and require multi‑step reasoning
• you prefer a Python SDK that works both locally and via a cloud API

When to avoid

• you rely on fast approximate similarity search for short texts
• your workload requires massive scale vector search across billions of embeddings
• you need out‑of‑the‑box support for non‑text modalities like images or audio

Facets

library · maturity active

llm-inference agent-framework artificial-intelligence large-language-models python cloud self-hosted vectorless tree-index reasoning-based document-indexing rag retrieval retrieval-augmented-generation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="VectifyAI/PageIndex")