Landing.ai
LandingAI, founded by Andrew Ng, builds production AI for visual and document understanding. Its flagship product is Agentic Document Extraction (ADE), which converts real-world documents — PDFs, images, Office files and spreadsheets — into structured, auditable data. ADE exposes a family of REST APIs: Parse (document to reading-order Markdown plus hierarchical blocks with per-block visual grounding), Extract (schema-driven field extraction with character-span grounding back into the source), Build Extract Schema, and the preview Classify, Section and Split APIs, each available synchronously or as asynchronous Jobs for large files. The current generation runs on the DPT-3 document pre-trained transformer. LandingAI also operates LandingLens, a computer-vision platform API for creating projects, uploading images, defining classes, training models and deploying them to LandingAI-hosted cloud endpoints, plus a Vision Tools API surfacing OWLv2, Florence-2, SAM2, CountGD, Depth Anything, PaddleOCR and other vision models as HTTP tools. The company ships official Python and TypeScript ADE libraries, a hosted remote MCP server, published Agent Skills, an llms.txt agent surface, US and EU data residency, and SOC 2 Type II, HIPAA and GDPR compliance with a zero data retention option.
Landing.ai publishes 4 APIs on the APIs.io network, including LandingAI Agentic Document Extraction (ADE) API v2, LandingAI Agentic Document Extraction (ADE) API v1, LandingAI Vision Tools API, and 1 more. Tagged areas include Company, Artificial Intelligence, Machine Learning, Document Extraction, and Document Processing.
Landing.ai’s developer surface includes documentation, API reference, getting-started guide, support, engineering blog, pricing, signup flow, and 27 more developer resources.
4 APIs
1 MCP Servers
0 Features
CompanyArtificial IntelligenceMachine LearningDocument ExtractionDocument ProcessingComputer VisionOCRData ExtractionAgentsDeveloper Tools
The current generation ADE API, powered by the DPT-3 document pre-trained transformer. Parse converts a PDF or image into reading-order Markdown plus a hierarchical block struct...
The v1 ADE surface covering Parse, Extract, Build Extract Schema and the preview Classify, Section and Split APIs, with synchronous and Jobs variants. Classify labels each page ...
The Vision Tools API exposes a catalog of computer-vision models as HTTP endpoints, including OWLv2, Florence-2, Florence2-SAM2, GLEE, CountGD, LOCA, Depth Anything V2, Depth Pr...
A RESTful API for using LandingLens programmatically, with predictable resource-oriented URLs and JSON-encoded responses. It covers creating projects, uploading images, creating...
aid: landingai
name: Landing.ai
description: 'LandingAI, founded by Andrew Ng, builds production AI for visual and document understanding. Its flagship product
is Agentic Document Extraction (ADE), which converts real-world documents — PDFs, images, Office files and spreadsheets
— into structured, auditable data. ADE exposes a family of REST APIs: Parse (document to reading-order Markdown plus hierarchical
blocks with per-block visual grounding), Extract (schema-driven field extraction with character-span grounding back into
the source), Build Extract Schema, and the preview Classify, Section and Split APIs, each available synchronously or as
asynchronous Jobs for large files. The current generation runs on the DPT-3 document pre-trained transformer. LandingAI
also operates LandingLens, a computer-vision platform API for creating projects, uploading images, defining classes, training
models and deploying them to LandingAI-hosted cloud endpoints, plus a Vision Tools API surfacing OWLv2, Florence-2, SAM2,
CountGD, Depth Anything, PaddleOCR and other vision models as HTTP tools. The company ships official Python and TypeScript
ADE libraries, a hosted remote MCP server, published Agent Skills, an llms.txt agent surface, US and EU data residency,
and SOC 2 Type II, HIPAA and GDPR compliance with a zero data retention option.'
image: https://landing.ai/favicon.ico
url: https://raw.githubusercontent.com/api-evangelist/landingai/refs/heads/main/apis.yml
x-type: company
x-source: vc-portfolio
x-backed-by:
- insight-partners
x-tier: enriched
specificationVersion: '0.20'
created: '2026-07-17'
modified: '2026-07-19'
tags:
- Company
- Artificial Intelligence
- Machine Learning
- Document Extraction
- Document Processing
- Computer Vision
- OCR
- Data Extraction
- Agents
- Developer Tools
maintainers:
- FN: Kin Lane
email: kin@apievangelist.com
- FN: APIs.json
email: info@apis.io
apis:
- aid: landingai:ade-v2
name: LandingAI Agentic Document Extraction (ADE) API v2
description: The current generation ADE API, powered by the DPT-3 document pre-trained transformer. Parse converts a PDF
or image into reading-order Markdown plus a hierarchical block structure with per-block grounding; Extract pulls schema
defined fields out of that Markdown with character-span grounding back into the source. Both run synchronously or as asynchronous
Jobs, which raise the ceilings to 1 GiB and 6,000 pages per PDF and support standard and priority service tiers.
humanURL: https://docs.landing.ai/dpt3/overview
baseURL: https://api.ade.landing.ai
tags:
- Document Extraction
- Parsing
- Structured Data
properties:
- type: OpenAPI
url: openapi/landingai-ade-v2-openapi-original.json
- type: Documentation
url: https://docs.landing.ai/dpt3/overview
- type: APIReference
url: https://docs.landing.ai/api-reference/parse/ade-parse
- type: GettingStarted
url: https://docs.landing.ai/dpt3/quickstart
- type: Overlay
url: overlays/landingai-ade-v2-overlay.yaml
- aid: landingai:ade-v1
name: LandingAI Agentic Document Extraction (ADE) API v1
description: The v1 ADE surface covering Parse, Extract, Build Extract Schema and the preview Classify, Section and Split
APIs, with synchronous and Jobs variants. Classify labels each page of a document by type, Section generates a hierarchical
table of contents, and Split separates a parsed document into classified sub-documents.
humanURL: https://docs.landing.ai/ade/ade-overview
baseURL: https://api.va.landing.ai
tags:
- Document Extraction
- Classification
- Sectioning
properties:
- type: OpenAPI
url: openapi/landingai-ade-v1-openapi-original.json
- type: Documentation
url: https://docs.landing.ai/ade/ade-overview
- type: APIReference
url: https://docs.landing.ai/api-reference/tools/ade-parse
- type: GettingStarted
url: https://docs.landing.ai/ade/ade-quickstart
- type: Overlay
url: overlays/landingai-ade-v1-overlay.yaml
- aid: landingai:vision-tools
name: LandingAI Vision Tools API
description: The Vision Tools API exposes a catalog of computer-vision models as HTTP endpoints, including OWLv2, Florence-2,
Florence2-SAM2, GLEE, CountGD, LOCA, Depth Anything V2, Depth Pro, PaddleOCR, pose detection, barcode and QR reading,
license-plate reading, activity recognition, NSFW classification, embeddings, and a vision-training job surface for custom
object detection.
humanURL: https://docs.landing.ai/
baseURL: https://api.va.landing.ai
tags:
- Computer Vision
- Object Detection
- Image Analysis
properties:
- type: OpenAPI
url: openapi/landingai-vision-tools-openapi-original.json
- type: Documentation
url: https://docs.landing.ai/
- type: Overlay
url: overlays/landingai-vision-tools-overlay.yaml
- aid: landingai:landinglens
name: LandingLens API
description: A RESTful API for using LandingLens programmatically, with predictable resource-oriented URLs and JSON-encoded
responses. It covers creating projects, uploading images, creating classes, auto-splitting datasets, training models and
reading training metrics and confusion matrices, managing dataset snapshots and exports, project metadata and tags, and
deploying trained models to LandingAI-hosted cloud endpoints.
humanURL: https://app.landing.ai
baseURL: https://api.landing.ai
tags:
- Computer Vision
- Model Training
- MLOps
properties:
- type: OpenAPI
url: openapi/landingai-landinglens-openapi-original.json
- type: Documentation
url: https://docs.landing.ai/landinglens-landingedge
- type: Overlay
url: overlays/landingai-landinglens-overlay.yaml
common:
- type: Website
url: https://landing.ai/
- type: DeveloperPortal
url: https://docs.landing.ai/
- type: Documentation
url: https://docs.landing.ai/
- type: APIReference
url: https://docs.landing.ai/api-reference/parse/ade-parse
- type: GettingStarted
url: https://docs.landing.ai/dpt3/quickstart
- type: Support
url: https://www.landing.ai/support
- type: Blog
url: https://www.landing.ai/resources
- type: GitHubOrganization
url: https://github.com/landing-ai
- type: Pricing
url: https://www.landing.ai/pricing-agentic-apis
- type: SignUp
url: https://ade.landing.ai/
- type: TermsOfService
url: https://www.landing.ai/terms-of-service
- type: PrivacyPolicy
url: https://www.landing.ai/privacy-policy
- type: StatusPage
url: https://status.landing.ai/
- type: TrustCenter
url: https://trust.landing.ai/
- type: Compliance
url: https://trust.landing.ai/
- type: Security
url: https://www.landing.ai/security-at-landingai
- type: Sandbox
url: sandbox/landingai-sandbox.yml
- type: Packages
url: packages/landingai-packages.yml
- type: SDKs
url: packages/landingai-packages.yml
- type: MCPServer
url: mcp/landingai-mcp.yml
- type: AgentSkill
url: skills/_index.yml
- type: LLMsTxt
url: llms/landingai-llms.txt
- type: Authentication
url: authentication/landingai-authentication.yml
- type: Conventions
url: conventions/landingai-conventions.yml
- type: ErrorCatalog
url: errors/landingai-problem-types.yml
- type: Lifecycle
url: lifecycle/landingai-lifecycle.yml
- type: Deprecation
url: https://docs.landing.ai/ade/ade-overview-legacy
- type: ChangeLog
url: changelog/landingai-changelog.yml
- type: Conformance
url: conformance/landingai-conformance.yml
- type: DataModel
url: data-model/landingai-data-model.yml
- type: RateLimits
url: rate-limits/landingai-rate-limits.yml
- type: Plans
url: plans/landingai-plans.yml
- type: WellKnown
url: well-known/landingai-well-known.yml
- type: DomainSecurity
url: security/landingai-domain-security.yml
x-enrichment:
date: '2026-07-19'
status: backfilled
pass: local-v1
note: backfilled from .gitignore signal + verified work evidence