Overview

(ADE) is a document intelligence platform that converts documents into reliable, structured data. Use the structured data to build retrieval-augmented generation (RAG) applications, power intelligent search systems, extract key information, and automate document processing. The Parse v2 API is powered by , our newest parsing model. is a complete rethink of document parsing for the era of AI agents and agentic workflows, designed around feedback from users running document-intensive workloads.

​

What DPT-3 Changes

The Parse v2 API is a redesign, not just a model update.

• Designed for agent systems. Built around how downstream AI agents consume document data: stable block ids, predictable hierarchy, and Markdown span pointers so you can map results back to source.
• Hierarchical layout model. The structured output is a nested hierarchy: pages, blocks on each page, lines within text blocks, and cells within tables.
• Line-level visual grounding. Bounding boxes per block, plus per-line boxes for text blocks, so you can highlight or extract content at the line level.
• Cleaner, more standardized Markdown. Structured figure elements with classification labels, bracketed attestation and scan-code labels, and standardized checkboxes.
• v2 endpoint and response shape. Calls go to the Parse v2 API. The response carries a hierarchical structure tree and a parallel grounding tree that mirrors it with spatial data. See Parse API Response.

​

Important to Know

A few things to keep in mind before integrating with the Parse v2 API.

​

Using the Parse v2 API Requires Code Changes

The endpoint, request shape, and response shape all differ from the Parse v1 API. Existing pipelines that call the v1 /v1/ade/parse endpoint need code changes to use the v2 /v2/parse endpoint. See the Migration Guide for the full diff.

​

Not Compatible with the Other ADE APIs

ADE Extract and ADE Section consume Parse output and expect the v1 response shape. If your workflow chains Parse with either of those, use the Parse v1 API for those workflows.

​

Not Available in the Python or TypeScript Libraries

The ade-python and ade-typescript libraries don’t yet include the v2 Parse endpoint. To use , call /v2/parse directly with cURL or your HTTP client of choice.

​

Only PDFs and Images Are Supported

The Parse v2 API supports PDFs and images only. For other file types, use the Parse v1 API. See Supported File Types for the full list.

​

Password-Protected Files Are Not Supported

The Parse v2 API does not support password-protected files. Decrypt the file and submit an unencrypted copy, or use the Parse v1 API.

​

Find What You Need

Pick a starting point based on what you want to do.