Frequently Asked Questions

Everything you need to know about Cargoffer OCR.

What document formats are supported?

PDF (invoices, CMRs, delivery notes), XLSX (price lists), EML (emails), PNG and JPG (scanned documents).

How accurate is the extraction?

We use LLM vision models (Gemini, DeepSeek, GLM) with a second verification pass. Accuracy depends on document quality but typically exceeds 95% for standard invoices.

Is there a free tier?

Yes. Free plan includes 10 pages/month. Pro plan ($49/mo) includes 1,000 pages. Enterprise ($499/mo) includes 100,000 pages.

Do you cache results?

Yes. Same document content (same SHA-256 hash) returns cached result across users. Soft delete preserves cache.

How does auto-model selection work?

PDF → Gemini (native vision). XLSX/EML → DeepSeek V4 Flash (fastest on structured data). Images → Gemini. Falls back to alternative models if primary fails.

Can I use my own API key?

The platform handles model selection internally. You only need our API key. No configuration required.

Is my data secure?

Documents are stored in MongoDB GridFS with 1-week retention. Cached results remain for 3 months if accessed. All transfers are TLS-encrypted.

How do I integrate with my ERP?

Use the REST API from any language. See our tutorials for SAP, Oracle, Dynamics 365, Odoo, and Sage.