Your documents never train a model, ours or a provider's
Every value we extract points back to the pixel it came from.
Invoices, receipts, purchase orders, delivery notes, scans and phone photos. It reads PDFs, images, and structured e-invoices. AI extracts the fields; every value links to the exact spot on the page it was read from.
Start free, no card required.
Live demo
Click a field, see the proof.
No hovering over a black box: select a value on the right and the exact region of the page it was read from lights up on the left.
- Extracted field
- Source value
- Highlighted on the page
LudwigstraΓe 22, MΓΌnchen
- 0 Documents used for AI training Ours or a provider's. Content sent to a provider is used only to produce your result.
- 100 Free pages to start One-time allowance on the free plan, no card required.
- 6 Languages, read natively The interface and extraction support English, German, French, Spanish, Polish, and Italian.
- 4 E-invoice formats read as XML DocAI reads Factur-X, ZUGFeRD, KSeF FA(3), and UBL 2.1 directly from XML, with exact values and no OCR.
Built for people who are accountable for the numbers.
Three commitments we hold to, and publish evidence for.
-
Never used for AI training.
Your documents train no model, ours or a provider's.
-
Every field links to its source.
Check any value against the exact spot on the page.
-
PDFs, photos, and e-invoices.
Scans and phone photos too, not just clean digital files.
Correct it once. DocAI remembers.
Every correction you make on a supplier's document is remembered and used as a hint on the next document from that vendor. You can see which hints were applied, and review or delete them at any time.
Invoice: March
IBAN PL61 1090 1014β¦ PL61 1090 2402β¦
corrected by you
Invoice: April
IBAN PL61 1090 2402β¦
hint applied
Got an e-invoice? We don't guess. We read the XML.
Factur-X, ZUGFeRD, KSeF FA(3), and UBL 2.1 carry an embedded structured layer. DocAI reads it directly: no OCR, no inference cost, no recognition error. AI runs only for documents without that layer.
See a KSeF FA(3) e-invoice in the live demo KSeF in Poland: how DocAI reads FA(3)
How it works
-
Step 01
Add documents
Upload PDFs, photos, or e-invoice XML, or send documents in by email or through the REST API.
-
Step 02
AI extracts, with evidence
Fields come back with a confidence score and a link to the exact spot on the page they were read from.
-
Step 03
Review, correct, export
Confirm or fix values in the review workspace, then export CSV, JSON, or XLSX, or receive results by webhook.
Use DocAI directly from your coding agent
DocAI connects to MCP-compatible coding agents, so they can discover the integration, inspect capabilities, and wire up structured document extraction in your project.
Tested with Codex, Claude Code and Gemini CLI. Same DocAI processing engine underneath as the REST API.
Simple plans, every feature included
Create a free account and the first 100 pages are on us, no card required. Paid plans refill your page balance to the plan's monthly amount each billing cycle. Unused pages carry over.
Free
$0 100 free pages to start
- 100 pages, one-time, yours until used
- 100 MB storage Β· 50 documents
Recommended
Pro
$29 / month
- 1,000 pages per month
- 5 GB storage Β· 2,000 documents
Business
$99 / month
- 4,000 pages per month
- 20 GB storage Β· 10,000 documents
Frequently asked questions
Where are my documents processed?
AI analysis sends up to the first three pages of a document to the AI provider configured for the deployment: OpenAI (United States) or OVH AI Endpoints (European Union). The privacy policy lists every subprocessor and what each one receives.
Do my documents train AI models?
No. Your documents and extracted data are never used to train AI models, ours or a provider's. Content sent to a provider is used only to produce your result.
How much of a long document does the AI read?
DocAI renders the first 5 pages and sends the first 3 to the model. Long documents are read in part, not in full. The same limits are stated on the pricing page.
What happens when I upload an e-invoice?
If the file carries a structured XML layer (Factur-X, ZUGFeRD, KSeF FA(3), UBL 2.1), DocAI reads the XML directly: exact values, no OCR, no AI inference. AI runs only for documents without that layer.
Is there an API?
Yes. A REST API with asynchronous jobs, email ingestion, and signed webhooks, on every plan. The API reference is public.
Is the free plan really free?
Yes. Create an account and the first 100 pages are free, one-time, no card required. Paid plans refill your page balance to the plan's monthly amount each billing cycle. Unused pages carry over.
Which languages and countries does DocAI work in?
The interface and extraction work in English, German, French, Spanish, Polish, and Italian. Documents in these languages are read natively. Regional variants add country-specific fields automatically: EU VAT details, US sales tax, India GST. You can also pick the region yourself before analysing.
See it on your own documents.
Upload one invoice. If the evidence links don't convince you in two minutes, nothing else will. The first 100 pages are free, no card required.