Skip to navigation
Get Started

Welcome to GroundX

GroundX turns PDFs, presentations, scanned forms, and websites into answers backed by source documents and structured JSON. Use it from application code or through an AI agent.

What GroundX Does

  • Document understanding: converts complex documents into clean, structured data for large language models (LLMs). It extracts fields, tables, and context that raw optical character recognition (OCR) misses.
  • End-to-end RAG: uses retrieval augmented generation to ingest documents into a searchable bucket and return answers with source references for an LLM.
  • Structured extraction: turns varied document formats into consistent structured JSON your application can use directly.
  • Document classification: sorts documents by type, intent, or workflow need to select the next extraction, review, or answer step.
  • Smart reports: runs a defined set of questions against documents and assembles the answers into a report with source references for review or further work.

Choose How You Want To Use GroundX

  • Agent Harness: let Claude, Codex, or another agent ingest documents, search, extract data, debug a stuck process, or explain GroundX. The harness provides GroundX instructions and workflows. It includes access to hosted Model Context Protocol (MCP) tools where supported.
  • SDKs/REST: use it when your application code calls GroundX directly. Call the Python or TypeScript SDK, or the REST API with X-API-Key authentication, from your own services.
  • On-prem: a deployment model for running GroundX in a controlled, security-conscious, or air-gapped environment.

Start Building