Overview
EU PII Safeguard is a production-ready multilingual PII detection model built specifically for European enterprises navigating GDPR compliance. Unlike generic NLP tools, it was trained and evaluated exclusively on European data across all 26 EU official languages, delivering consistent 95%+ F1 scores from German and French to Maltese and Irish.
WHAT IT DETECTS - 42 PII Entity Types: Personal identifiers (names, age, gender, ethnicity), contact information (email, phone, address, postal code), financial data (IBAN, credit card, account numbers, salary), government IDs (national ID, passport, driver's license, tax ID), health information (medical conditions, health insurance IDs), and digital identifiers (IP address, MAC address, username, password).
WHO IT'S FOR:
- Legal and compliance teams automating GDPR data audits
- Data engineering teams anonymizing datasets before ML training
- SaaS platforms handling user-generated content across EU markets
- Healthcare and financial institutions with strict data protection obligations
- Document processing pipelines (contracts, emails, support tickets)
WHY EU PII SAFEGUARD:
- 97.02% Global F1, best-in-class for European multilingual PII
- Made in Europe by Tabularis AI, a University of Tuebingen spin-off
- On-premises deployment via SageMaker, your data never leaves AWS
- No internet access required at inference time, fully air-gapped capable
- Supports both real-time endpoints and batch transform jobs
- Simple REST API: send text, receive detected entities in JSON
Highlights
- 97% F1 accuracy across all 26 EU official languages - consistent performance from German and French to Maltese, Irish, and Bulgarian with 42 detected PII entity types including IBANs, national IDs, and medical data.
- GDPR-compliant by design: deploy entirely within your AWS VPC with no outbound internet calls. Your sensitive data never leaves your infrastructure - ideal for regulated industries including healthcare, finance, and legal.
- Production-ready in minutes: deploy as a SageMaker real-time endpoint or batch transform job. Standard JSON API compatible with LangChain, AWS Lambda, and any Python or REST-capable application.
Details
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.m5.xlarge Inference (Batch) Recommended | Model inference on the ml.m5.xlarge instance type, batch mode | $0.15 |
ml.m5.xlarge Inference (Real-Time) Recommended | Model inference on the ml.m5.xlarge instance type, real-time mode | $0.30 |
ml.m5.2xlarge Inference (Batch) | Model inference on the ml.m5.2xlarge instance type, batch mode | $0.25 |
ml.c5.2xlarge Inference (Batch) | Model inference on the ml.c5.2xlarge instance type, batch mode | $0.20 |
ml.g4dn.xlarge Inference (Batch) | Model inference on the ml.g4dn.xlarge instance type, batch mode | $0.40 |
ml.m5.2xlarge Inference (Real-Time) | Model inference on the ml.m5.2xlarge instance type, real-time mode | $0.50 |
ml.c5.2xlarge Inference (Real-Time) | Model inference on the ml.c5.2xlarge instance type, real-time mode | $0.40 |
ml.g4dn.xlarge Inference (Real-Time) | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $0.80 |
Vendor refund policy
We offer a 3-day refund for new subscribers who experience technical issues preventing the product from functioning as described. To request a refund, contact info@tabularis.ai within 3 days of purchase with your AWS account ID and a description of the issue. Refund requests are reviewed within 2 business days. No refunds are issued for usage beyond 3 days or for infrastructure costs charged by AWS.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Version release notes
Initial release of EU PII Safeguard. Multilingual PII detection model supporting 26 European languages and 42 entity types. Based on XLM-RoBERTa-large, fine-tuned on European PII data. Achieves 97% F1 score. Supports both single-text and batch inference via JSON API.
Additional details
Inputs
- Summary
The model accepts JSON input with Content-Type: application/json.
Single text request: {"text": "Your text string here"}
Batch request (multiple texts in one call): {"texts": ["First text", "Second text", "Third text"]}
The input text can be in any of the 26 supported European languages. Maximum input length is 256 tokens (approximately 200 words). Longer texts are automatically truncated.
Supported languages: English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Swedish, Danish, Norwegian, Finnish, Czech, Slovak, Hungarian, Romanian, Bulgarian, Croatian, Slovenian, Lithuanian, Latvian, Estonian, Greek, Irish, Maltese, and Luxembourgish.
- Limitations for input type
- Maximum token length: 256 tokens per text. Texts exceeding this limit are truncated. Content-Type must be application/json. Each text should be a single string (not nested objects).
- Input MIME type
- application/json
Support
Vendor support
Email: info@tabularis.ai | Response time: 1-3 business days Website: https://tabularis.ai Documentation: https://huggingface.co/tabularisai/eu-pii-safeguard
Tabularis AI provides email support for all subscribers. Support covers deployment issues, API usage questions, and bug reports. Enterprise support contracts with guaranteed SLAs and dedicated response times are available on request.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.