Google Document AI

"A document processing and understanding platform that takes unstructured data from documents and transforms it into structured data, making it easier to understand, analyze, and consume." [1]

cloud.google.com/document-ai · By Google · Agent JSON · Suggest an edit · Last verified 2026-06-14 · Source confidence: high

Google Document AI is a REST API from Google Cloud that transforms unstructured documents into structured data, covering OCR, data extraction from invoices, receipts, and forms, identity document verification, and custom trained extraction models. Pricing is usage-based at $0.02 per 1,000 pages with self-serve signup and no sales call required. The API ships official SDKs for eight languages including Python, Java, Node.js, and Go, and is available across eight regions including US, EU, and Asia-Pacific endpoints. It carries SOC 2 Type 2, ISO 27001, HIPAA, GDPR, and PCI DSS compliance certifications.

Best for / Avoid if

Best for: Teams needing broad API coverage out of the box; Cost-sensitive teams - low, transparent entry price

Avoid if: You want to try it free before paying

Scores

  • 30 / 100
    Agent friendliness
  • 85 / 100
    Pricing transparency
  • 60 / 100
    Setup speed
  • 15 / 100
    Docs quality
  • 85 / 100
    Procurement ease
  • 100 / 100
    Trust readiness

Scores are computed deterministically from this profile's published, sourced fields (pricing, compliance, capabilities, docs and developer-surface signals) - not from reviews or paid placement. Each axis is 0-100; an unknown signal scores 0 for that axis. Procurement ease is the inverse of buying friction (higher = easier to adopt).

Pricing & procurement

Pricing model
Usage-based [2]
Published pricing
Yes [3]
Free tier
No [4]
Self-serve signup
Yes [5]
Requires sales call
No
Enterprise plan
No
Published prices
PlanItemPerAmountSource
1–5,000,000 pages/monthEnterprise Document OCR Processor1,000 pages$1.5source
5,000,001+ pages/monthEnterprise Document OCR Processor1,000 pages$0.6source
OCR add-ons1,000 pages$6source
1–1,000,000 pages/monthCustom Extractor1,000 pages$30source
1,000,001+ pages/monthCustom Extractor1,000 pages$20source
1–1,000,000 pages/monthForm Parser1,000 pages$30source
1,000,001+ pages/monthForm Parser1,000 pages$20source
1–1,000,000 pages/monthLayout Parser (includes initial chunking)1,000 pages$10source
1,000,001+ pages/monthLayout Parser (includes initial chunking)1,000 pages$10source
Re-chunking parsed documents1,000 pages$0.02source
1–1,000,000 pages/monthCustom Splitter1,000 pages$5source
1,000,001+ pages/monthCustom Splitter1,000 pages$3source
1–1,000,000 pages/monthCustom Classifier1,000 pages$5source
1,000,001+ pages/monthCustom Classifier1,000 pages$3source
1–1,000,000 pages/monthSummarizer1,000 pages$25source
1,000,001+ pages/monthSummarizer1,000 pages$25source
Invoice Parser10 pages$0.1source
Expense Parser10 pages$0.1source
Utility Parser10 pages$0.1source
Procurement Document Splitter & Classifierclassified document$0.05source
Bank Statement Parserclassified document$0.75source
Pay Slip Parserclassified document$0.3source
W2 Parserclassified document$0.3source
Lending Document Splitter & Classifierclassified document$0.05source
US Driver License Parserdocument$0.1source
US Passport Parserdocument$0.1source
Identity Document Proofingdocument$0.1source
Provisioned (Reserved Capacity)Reserved capacity (extra page-per-minute slot)page-per-minute per month$300source
Custom processor version hostinghour per deployed processor version$0.05source

Capabilities

  • Receipts / invoices
  • ID documents
  • Bank statements
  • Table extraction
  • Custom models
  • LLM / RAG-ready output
Supported actions
process (synchronous document processing), batchProcess (asynchronous batch document processing), create processor, delete processor, enable processor, disable processor, list processors, get processor, setDefaultProcessorVersion, deploy processor version, undeploy processor version, train processor version, delete processor version, list processor versions, evaluateProcessorVersion, fetchProcessorTypes, list processor types, get processor type, cancel operation, get operation, list operations, reviewDocument (human review, deprecated), importDocuments (dataset), listDocuments (dataset), getDocument (dataset), batchDeleteDocuments (dataset), updateDatasetSchema, getDatasetSchema, create schema, delete schema, get schema, list schemas, patch schema, create schema version, generate schema version, get schema version, list schema versions, patch schema version [6]
Regions
us (United States multi-region), eu (European Union multi-region), asia-south1 (Mumbai), asia-southeast1 (Singapore), australia-southeast1 (Sydney), europe-west2 (London), europe-west3 (Frankfurt), northamerica-northeast1 (Montréal) [7]
Languages
English, French, German, Spanish, Italian, Portuguese, Dutch, Polish, Russian, Ukrainian, Chinese (Simplified), Chinese (Traditional), Japanese, Korean, Hindi, Bengali, Thai, Vietnamese, Arabic, Persian, Hebrew, Afrikaans, Albanian, Armenian, Bulgarian, Croatian, Czech, Danish, Estonian, Finnish, Greek, Hungarian, Icelandic, Indonesian, Latvian, Lithuanian, Macedonian, Malay, Marathi, Nepali, Norwegian, Romanian, Serbian, Slovak, Slovenian, Swedish, Turkish, Yiddish, 200+ languages supported for Enterprise Document OCR and Form Parser [8]
Input types
application/pdf, image/jpeg, image/png, image/tiff, image/gif, image/bmp, image/webp, text/html (Layout Parser only), application/vnd.openxmlformats-officedocument.wordprocessingml.document (Layout Parser only), application/vnd.openxmlformats-officedocument.presentationml.presentation (Layout Parser only), application/vnd.openxmlformats-officedocument.spreadsheetml.sheet (Layout Parser only) [9]
Output types
JSON [10]
Webhooks
No [11]
Sandbox / test mode
No [12]
SDK languages
Python, Java, Node.js, Go, C#, PHP, Ruby, C++ [13]
MCP server
No

Trust & compliance

SOC 2
SOC 2 Type II [14]
HIPAA
Yes [15]
GDPR
Yes [16]
ISO 27001
Yes [17]
PCI DSS
Yes [18]
Published SLA
Yes [19]
Rate limits
Online process requests per minute (US or EU multi-region): 120 per project per processor type; Online process requests per minute (single region): 6 per project per processor type; Concurrent batch requests (US or EU): 5 per project; Concurrent batch requests (single region): 5 per project; Base requests per minute: 1,800 per user; Default pages per minute (Best Effort, base processors): 120 PPM; Default pages per minute (Best Effort, Pro processors): 60 PPM; Concurrent training requests: 1 per project per region; Deployed custom processor versions: 5 per project per region; Deployed generative processor versions: 100 per project per custom extraction processor. [20]
Known restrictions
HTML and OOXML (DOCX, PPTX, XLSX) input supported only with Layout Parser, Custom Splitter only supports PDF, TIFF, TIF, and GIF input, Recommended minimum image resolution: 200 dpi; 300+ dpi for best results, Utility Parser and Procurement Document Splitter & Classifier require form submission for access (limited access), Custom processor hosting billed at $0.05 per hour per deployed processor version regardless of usage, Provisioned capacity reservation: $300 USD per extra page-per-minute per month, 4xx/5xx error responses are not billed, Only English officially supported for Custom Extractor when using generative AI extraction, Batch processing results stored in Cloud Storage (not returned inline), DOCX page unit: up to 3,000 characters = 1 page; XLSX: each tab = 1 page; PPTX: each slide = 1 page; HTML/parsed docs: up to 3,000 characters = 1 page [21]

Developer surface

Docs rendering: static

Integration

API style
rest
Base URL
https://documentai.googleapis.com/v1
Version
v1
Versioning
url
Stability
ga
Auth methods
oauth2
Idempotency keys
No
Error format
google.rpc.Status
Rate limit
600 / minute

SDKs

  • Python google-cloud-documentai · repo
  • Java com.google.cloud:google-cloud-document-ai · repo
  • Node.js @google-cloud/documentai · repo
  • Go cloud.google.com/go/documentai · repo
  • C# Google.Cloud.DocumentAI.V1 · repo
  • PHP google/cloud-document-ai · repo
  • Ruby google-cloud-document_ai · repo
  • C++ · repo

Adoption & maturity

Launched
2020-10-16
GA
2021-03-31
Notable customers
Covered California, Gogolook

Other OCR & Document Parsing APIs

  • Amazon Textract

    "Automatically extract printed text, handwriting, layout elements, and data from any document"

    Usage · public pricing · self-serve

  • Veryfi

    "Documents into Data - securely, in seconds"

    Hybrid · free tier · public pricing · self-serve

  • Azure AI Document Intelligence

    "Azure Document Intelligence in Foundry Tools is a machine-learning based OCR and intelligent document processing service to automate extraction of key data from forms and documents."

    Usage · free tier · public pricing · self-serve

  • Nanonets

    "AI Agents for Enterprise Data Processing."

    Usage · public pricing · self-serve

  • Extend

    "Turn documents into high quality data"

    Hybrid · public pricing · self-serve

  • Mindee

    "Turn your document data into structured JSON with high-reliability. Zero model training required."

    Hybrid · public pricing · self-serve

Google Document AI alternatives · Google Document AI vs Amazon Textract · All OCR & Document Parsing APIs APIs

References

Change history

Every field change, who made it, and when - from our audited data pipeline and editors.

  1. 2026-06-15 Score Agent Friendliness: 1030
  2. 2026-06-14 Has Structured Data: (none)Yes
  3. 2026-06-14 Robots Allows Agents: (none)Yes
  4. 2026-06-14 Capabilities: {}{"tables":true,"id_documents":true,"custom_models":true,"agentic_output":true,"…
  5. 2026-06-14 Summary Md: (none)Google Document AI is a REST API from Google Cloud that transforms unstructured…
  6. 2026-06-14 Score Docs Quality: 015
  7. 2026-06-14 Docs URL: (none)https://docs.cloud.google.com
  8. 2026-06-14 Rendering: (none)static
  9. 2026-06-14 Status Page URL: (none)https://status.cloud.google.com
  10. 2026-06-14 Llms Txt Present: (none)No
  11. 2026-06-14 Scoring Methodology: (none)Scores are computed deterministically from this profile's published, sourced fi…
  12. 2026-06-14 Avoid If: (none)You want to try it free before paying
  13. 2026-06-14 Score Agent Friendliness: (none)10
  14. 2026-06-14 Score Pricing Transparency: (none)85
  15. 2026-06-14 Score Setup Speed: (none)60
  16. 2026-06-14 Score Docs Quality: (none)0
  17. 2026-06-14 Score Procurement Friction: (none)85
  18. 2026-06-14 Score Trust Readiness: (none)100
  19. 2026-06-14 Best For: (none)Teams needing broad API coverage out of the box, Cost-sensitive teams - low, tr…
  20. 2026-06-14 SDK Packages: Python, Java, Node.js, Go, C#, PHP, Ruby, C++Python, Java, Node.js, Go, C#, PHP, Ruby, C++
  21. 2026-06-14 Last Verified At: 2026-06-13T00:00:00.000Z2026-06-14T00:00:00.000Z
  22. 2026-06-14 Starting Price Usd: 0.020.02
  23. 2026-06-13 Requires Sales Call: set to No
  24. 2026-06-13 Enterprise Plan Available: set to No
  25. 2026-06-13 GDPR: set to Yes
  26. 2026-06-13 ISO 27001: set to Yes
  27. 2026-06-13 PCI DSS: set to Yes
  28. 2026-06-13 SLA Published: set to Yes
  29. 2026-06-13 SLA URL: set to https://cloud.google.com/document-ai/sla
  30. 2026-06-13 Data Retention Policy URL: set to https://cloud.google.com/document-ai/docs/security
  31. 2026-06-13 Documented Rate Limits: set to Online process requests per minute (US or EU multi-region): 120 per project per…
  32. 2026-06-13 Rate Limit Requests: set to 600
  33. 2026-06-13 Rate Limit Window: set to minute
  34. 2026-06-13 Known Restrictions: set to HTML and OOXML (DOCX, PPTX, XLSX) input supported only with Layout Parser, Cust…
  35. 2026-06-13 Auth Methods: set to oauth2
  36. 2026-06-13 Auth Docs URL: set to https://cloud.google.com/docs/authentication/provide-credentials-adc
  37. 2026-06-13 API Style: set to rest
  38. 2026-06-13 Base URL: set to https://documentai.googleapis.com/v1
  39. 2026-06-13 API Version: set to v1
  40. 2026-06-13 Versioning Scheme: set to url
  41. 2026-06-13 Stability: set to ga
  42. 2026-06-13 Deprecation Policy URL: set to https://cloud.google.com/document-ai/docs/deprecation
  43. 2026-06-13 Quickstart URL: set to https://cloud.google.com/document-ai/docs/process-documents-client-libraries
  44. 2026-06-13 Idempotency Supported: set to No
  45. 2026-06-13 Error Format: set to google.rpc.Status
  46. 2026-06-13 Requires Verification: set to Yes
  47. 2026-06-13 Starting Price Usd: set to 0.02
  48. 2026-06-13 Price Basis: set to 1,000 pages
  49. 2026-06-13 Slug: set to google-document-ai
  50. 2026-06-13 GA Date: set to 2021-03-31

Suggest an edit / leave a review

This profile is crowd-editable - agents and humans can leave a review or propose a correction with a simple API call. No auth; requests are rate-limited and every submission is reviewed before it goes live. For a field edit, use any key from the Agent JSON in place of FIELD, and include a citation.

Leave a review or comment

curl -X POST https://apio.sh/api/feedback/google-document-ai \
  -H 'Content-Type: application/json' \
  -d '{"kind":"review","rating":5,"body":"Your experience with this API…"}'

Suggest a correction to a field (cite a source)

curl -X POST https://apio.sh/api/suggest/google-document-ai/FIELD \
  -H 'Content-Type: application/json' \
  -d '{"value":"corrected value","citations":[{"url":"https://source.example/page","excerpt":"supporting quote"}],"note":"what changed and why"}'

All the ways to contribute →