PageIndex

What it does

PageIndex is a document retrieval layer for AI applications. It builds a tree index of a document, then lets an LLM or agent search that structure for relevant information. Developers can submit a PDF, query it, and receive streamed answers with citations. Local citations identify pages; PageIndex Cloud can also provide block-level references.

The product is for developers building document-question-answering applications and agents, as well as people who want to ask questions of long documents in the Chat App. Options include an open-source local SDK, managed Cloud indexing and retrieval, and Enterprise deployment in a customer’s private cloud. The documented SDK workflow is to install the package, configure a PageIndex API key for Cloud and credentials for an LLM, then index and query a document. Local use runs on the user’s machine with their own model key.

PageIndex’s Cloud workflow uses a separately supplied LLM, whose provider is paid directly. Cloud charges for indexing and monthly active-page storage; retrieval on active pages has no PageIndex fee. Local software is free and open source, but model and compute costs are not specified. Scanned-file and image-understanding support is listed for Cloud; local is described as handling text-based PDFs without OCR.

What it can do

• Builds a tree index of document structure
• Retrieves relevant information from indexed documents
• Supports LLM- or agent-based document question answering
• Streams answers through the SDK
• Provides page citations locally and block-level citations in Cloud
• Cloud supports scanned files and image understanding

Works with

• LLMs and agents
• OpenAI models (shown as an example)

Before you choose

• The documented Cloud workflow requires a separate LLM and API key; users pay their LLM provider directly.
• Local citations stop at the page; block-level references are a PageIndex Cloud feature.
• Local is described for text-based PDFs without OCR; scanned-file and image-understanding support is listed for Cloud.

Listed under

• Data
• Ask your data
• Developers
• Document AI
• Enterprise
• Long documents
• Open source
• PDF

Get started

The developer page links to the Developer Dashboard to create an API key and begin on Cloud.

1. Install the SDK with `pip install -U pageindex` for the documented developer workflow.
2. For Cloud, create a PageIndex API key in the Developer Dashboard and prepare an LLM model and its provider credentials.
3. Submit and index a document, then query it through the SDK or an LLM/agent integration.

Cloud indexing is charged once per page indexed; active pages are billed monthly after the first 1,000 free pages. Retrieval on active pages has no PageIndex fee; users pay their LLM provider directly. Local is free and open source, but model and compute costs are not specified.

Source: pageindex.ai/developer, checked 1 Nov 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 1 Nov 2026. Each source supports part of the overview; availability and prices can change.

• PageIndex Developer ↗pageindex.ai
• PageIndex ↗vectify.ai
• PageIndex ↗pageindex.ai

Company website

pageindex.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 1 Nov 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.