DocParse
Platform
Documents
Custom fields
Define your own schema. Extract from anything.
Invoices
Line items, totals, taxes, supplier.
Resumes
Skills, experience, education, contacts.
Contracts
Parties, dates, amounts, clauses.
Receipts
Merchant, items, taxes, currency.
Bank statements
Account summary, transactions, balances.
Bill of lading
Shipper, consignee, cargo lines, terms.
Purchase orders
PO number, line items, delivery terms.
Emails
Sender, subject, body, attachments.
Browse all documents
Pricing
Resources
The basics
Document Information Extraction
What it is and how DocParse approaches it.
Automated Document Recognition
How multi-modal AI replaces legacy OCR.
Intelligent Data Automation
API, webhooks, and Zapier in practice.
Business Intelligence Data Mining
Turn documents into rows for your warehouse.
Help
API documentation
Authentication, endpoints, webhooks.
FAQ
Comprehensive product Q&A.
Contact
Email reply within one business day.
Contact
Sign in
Start extracting
You might ask
Frequently asked
questions.
Everything teams want to know before they sign up — straight answers, no fluff. If you can’t find what you need here,
get in touch
.
On this page
Platform basics
Documents & templates
File formats & limits
API, webhooks & integrations
Pricing & billing
Security & data
Support & troubleshooting
Full docs →
Platform basics
What is DocParse?
DocParse is an AI document processing platform with two modules — Data Extraction (turn unstructured documents into JSON) and Document Classification (sort incoming documents into categories). Both run on frontier multi-modal models, reachable through a REST API, signed webhooks, and a native Zapier app.
How do I get started?
Create an account in the dashboard, verify your email, and you'll see 100 free pages added immediately. Pick a built-in template, upload a file, and inspect the JSON output. From there, generate an API key or configure a webhook endpoint to wire DocParse into your own systems.
Do I need to be a developer to use it?
No. The dashboard handles everything end-to-end — defining fields, uploading files, viewing results, exporting JSON or CSV. For automation, the native Zapier app lets you trigger DocParse from Gmail, Drive, Slack, Sheets, and 6,000+ other apps without writing code.
Documents & templates
Which document types are supported out of the box?
Eight built-in templates: invoices, receipts, contracts, resumes, bank checks, business cards, bills of lading, and emails. Plus a custom-fields template for anything else — define your own fields by name, type, and description.
How do I define a custom schema?
Pick the custom template, then add fields one at a time: name, type (string, number, date, boolean, array, object), and an optional description that helps the model. Mark which fields are required. You can edit and re-run extractions without re-uploading the file.
What document options can I toggle?
Seven flags: tables, charts, checkboxes, handwritten, multi-page, split-PDF, and specific-pages. Each one tells the model to look for that specific structure or apply that processing rule, so you get richer output without rewriting your schema.
Can I classify documents before extracting?
Yes — that's the Document Classification module. Define categories with a name, description, and keywords. Optionally link each category to a target extraction so DocParse classifies the file first, then routes it through the right extraction template.
File formats & limits
Which file formats are accepted?
PDF, PNG, JPG / JPEG, WEBP, DOCX, and plain text. Phones photos, scanned PDFs, and digitally-generated PDFs all go through the same pipeline. DOCX is converted to text before extraction.
What are the size limits?
25 MB per file, 30 files per batch. Multi-page PDFs are fine within the 25 MB envelope — every rendered page counts toward your page balance.
Which languages are supported?
Any language the underlying multi-modal model handles. The default is multi-lingual so mixed-script documents work without extra configuration. You can also pin a specific language per extraction if you want the model to bias for it. Handwriting and right-to-left scripts are supported.
API, webhooks & integrations
How does the REST API work?
Plain HTTP with bearer-token authentication. Create an extraction (POST /api/v1/createExtraction), upload files to it (POST /api/v1/uploadFiles), and either poll for results (GET /api/v1/getBatchResults) or receive a signed webhook when the batch finishes.
Are there native SDKs?
Not yet. The API is plain REST + JSON, so any language with an HTTP client works — we publish copy-pasteable curl, Python (requests), and Node.js (fetch) examples in the docs. Native SDKs may follow once demand justifies them.
How are outbound webhooks secured?
Following the Standard Webhooks spec: every delivery includes `webhook-id`, `webhook-timestamp`, and `webhook-signature: v1,<sig>` headers. The signature is HMAC-SHA256 over `{id}.{timestamp}.{body}` using your per-endpoint secret. Timestamps outside a 5-minute window are rejected to prevent replay attacks.
Do you support Zapier?
Yes — DocParse ships a native Zapier app. Triggers: New File Processed. Actions: Upload File, Create Extraction. Search: Find Extraction by Name. Wires DocParse into Gmail, Drive, Dropbox, Slack, Sheets, Airtable, HubSpot, Salesforce, and 6,000+ other apps with no code.
Do you have Make.com or n8n integrations?
Both are supported through generic HTTP webhook nodes — point them at the DocParse webhook payload or call the REST API directly. Native nodes for Make.com and n8n aren't on the roadmap yet.
Pricing & billing
How does pricing work?
Every account gets 100 pages every month, free, forever. Beyond that, choose pay-as-you-go (buy a page pack once; pages never expire) or a monthly subscription at a lower per-page rate. Nine volume tiers from 100 to 50,000 pages, in either USD or INR.
What counts as a page?
One rendered page of any document. A 10-page PDF costs 10 pages. A single image or one-page DOCX is one page. Failed extractions (e.g. invalid file) don't count.
Can I pay in INR?
Yes. The checkout flow includes a currency switcher — pick USD or INR before paying. Payments are processed by Dodo Payments as our merchant of record.
Do unused pages roll over?
PAYG pages never expire — they sit on your account until you use them. Subscription pages reset at the start of each billing cycle. The free 100 pages reset on the 1st of every month.
Can I switch between PAYG and subscription?
Yes, anytime, no contract. Subscriptions can be cancelled or changed from the customer portal in the dashboard.
I need more than 50,000 pages a month — what do I do?
Drop us a line through the contact form. We'll work out a per-page rate and billing cycle that matches your actual volume, in USD or INR.
Security & data
Are my documents used to train models?
No. Your documents go to the model, the structured output comes back, and that's the end of it. Nothing is retained for training — ours or any third-party provider's.
How are API keys stored?
Hashed with SHA-256. We can't see the key after creation — only its prefix (the first 10 characters) is stored in plaintext so you can identify it in the dashboard. Revoke or rotate any key instantly, with per-key usage logs.
Is data encrypted?
Yes, in transit. Every request to the API, dashboard, and webhook endpoints uses TLS. Files at rest live in Supabase Storage, which encrypts at rest by default. File downloads use short-lived signed URLs that expire after one hour.
Are you SOC 2 / ISO 27001 / HIPAA certified?
No — we've not pursued external certifications yet. We list the concrete security controls we do implement (TLS, HMAC-signed webhooks, hashed API keys, signed download URLs, no training on customer data) so you can evaluate without relying on a logo.
Support & troubleshooting
Where can I find documentation?
The full API reference, webhook spec, and how-to guides live at our docs site (linked in the nav). It covers authentication, endpoint reference for extraction and classification, polling vs webhook patterns, and the Zapier integration.
How do I get help?
All plans include email support. The contact form on this site is the fastest way to reach us — we reply within one business day.
Where do I report a bug or request a feature?
Use the contact form to describe what you saw vs what you expected. Screenshots and the request ID (visible in the dashboard and in webhook deliveries) help us trace the issue quickly.
Still have questions?
Email support is on every plan. We reply within one business day.
Try it free
Contact support