Features
Support for 75+ languages
84 trained pipelines for 25 languages
Named entity recognition (NER)
Part-of-speech (POS) tagging
Multi-task learning with pretrained transformers like BERT
Config-driven training with no hidden defaults
spacy-llm package for LLM integration
Built-in visualizers for syntax and NER
Custom model support via PyTorch, TensorFlow
OCR and data extraction from invoices, receipts, IDs
Document classification and routing
Data anonymization and redaction
Document verification with fraud detection
Identity verification with liveness and NFC
Proof of income and age verification
Human-In-The-Loop workflow builder
Prompt builder for AI extraction templates
OCR API with multi-language support
Document verification API
Identity verification API and SDK
E-invoicing via Peppol network
Spend management (expense reports, approvals)
3-way matching (PO, invoice, receipt)
CO2 registration for expense tracking