Parse any document to make it computable
Meet Parse. Parse allows you to:
- Convert documents into accurate machine-readable formats.
- Preserve the document hierarchy and layout.
- Obtain precise citations for each content block and table cell.
Common Parsing Challenges
Developing and maintaining a template for each document layout is labor intensive, error-prone and not scalable.
Processes developed for clean PDFs cannot handle blur, distortion and rotation in document scans and photographs.
Users complete form fields and mark checkboxes in inconsistent ways - not always staying within the lines!
Visual media such as graphs, flowcharts, illustrations and diagrams is either ignored or requires separate logic.
Not all document parsers created equal.
High-quality parsing goes beyond basic OCR, which only turns pixels into text. To make a document truly computable, the parser must understand what each section means and how sections relate. ADE Parse adds a cognitive layer that reasons over documents — returning spatial relationships, visual grounding, and parsing confidences reliably at scale.
Accuracy on complex docs
Proven on real-world layouts, complex tables, and multi-page documents delivering consistent results in production, not just in benchmarks.
Results come with proof
Verify parsed results with page numbers and precise coordinates for each chunk. Confidence scoring surfaces results that may need human review.
Unmatched speed & scale
Eliminate processing bottlenecks and scale effortlessly. ADE handles thousands of pages per minute.
What you get
- No fine-tuning or templates required
- Agentic system that plans and executes
- Developer-friendly APIs
- Returns clean Markdown and JSON
- Layout and reading order detection
- Supports many filetypes
- Works across languages
- Handles files up to 1 GB
- Detects signatures and handwriting
- Recognizes checkboxes on forms
- Reconstructs messy tables
- Enriches diagrams and figures
- Assigns confidence scores
- Returns bounding box coordinates
- Fully managed SaaS or VPC
- Real-time or batch processing
Built for complex, real-world documents.
Parse is made for complex, real-world documents. It can handle your massive tables, scientific charts, handwritten forms, poor quality faxes, and much more.
Read the Docs for Parse →
Prototype in the UI. Scale with the API.
Use the Agentic Document Extraction Visual Playground for rapid prototyping and experimentation. Then transition seamlessly to the REST API, Python library, or TypeScript library for your production workloads.
Provide our Agent Skills directly to agentic coding assistants to accelerate your development.
curl -X POST 'https://api.ade.landing.ai/v2/parse' \
-H 'Authorization: Bearer YOUR_API_KEY' \
-F 'document=@document.pdf' \
-F 'model=dpt-3-pro-latest'from landingai_ade import LandingAIADE
client = LandingAIADE()
# Replace with your file path
response = client.parse(
document=Path("/path/to/file/document"),
model="dpt-2-latest",
save_to="./output"
)
print(response.chunks)
Frequently asked questions
A Document Pre-Trained Transformer (DPT) is the model that powers the parsing capabilities of the ADE Parsing APIs. The DPT identifies document layouts and chunks, then generates descriptive explanations (captions) for those chunks. All foundational DPT models are developed in-house by Landing AI by applying deep learning architectures. The T in DPT indicates that the model uses a Transformer architecture.
