PII Anonymization for the Agentic Era

A self-hosted API for detecting and redacting sensitive data in text and LLM chats. Supports 6 languages across 15 countries.

Raw Input

Hello, my name is Sarah Smith and I am a resident of London. You can reach me at sarah.smith@email.com or call my mobile at +44 7700 900123 regarding my account number 883920.

Safe Output

Hello, my name is <NAME> and I am a resident of <LOCATION>. You can reach me at <EMAIL> or call my mobile at <PHONE> regarding my account number <ACCOUNT>.

Deploy on your infrastructure in minutes

Unlimited usage licensing with no per-token fees. Simple contracts, no lengthy procurement.

Designed for Regulated Sectors

Built for compliance across regulated industries such as financial services, legal and insurance.

OpenAI Chat Support

Easily create guardrails with native support for the OpenAI chat completions format. Benefit from improved accuracy and speed via intelligent context handling.

Powerful Models

Built with the latest transformer technology, ensuring accurate detection on real-world data that might contain RAG ingestion artifacts and internet slang.

60+ Localized Entity Types

Languages don’t follow borders. PII Eraser supports dozens of country-specific entity types such as Australian TFNs and Austrian Firmenbuchnummern (FN).

Air-gapped Deployment

Runs entirely within your infrastructure (VPC/On-prem) in a single hardened container. No external connections required.

1M+ Token Context

Process entire documents and chat histories, with linear compute scaling and no accuracy degradation.

Realtime Power

>5000 tokens/sec on a 8 vCPU instance, with optimizations for the latest x86 CPUs like AMD EPYC Turin.

Minimal Code. Maximum Protection.

We handle the complexity, from mixed-language inputs to large documents with ingestion artifacts.

Python Input

import requests

# Detect sensitive data instances
response = requests.post(
    "http://localhost:8000/text/detect",
    json={
        "text": ["Hallo Matthias"],
    }
)

import requests

# Replace sensitive data with entity types
response = requests.post(
    "http://localhost:8000/text/transform",
    json={
        "text": ["I live at Berlinerstr 34, Magdeburg"],
        "operator": "redact", 
    }
)

import requests

# Mask characters with asterisks
response = requests.post(
    "http://localhost:8000/text/transform",
    json={
        "text": ["My Nino is QQ689643C"],
        "operator": "mask", 
    }
)

import requests

# Cryptographic hashing for analytics
response = requests.post(
    "http://localhost:8000/text/transform",
    json={
        "text": ["Medicare no. 3278851195"],
        "operator": "hash", 
    }
)

Output JSON

response.json()

{
  "entities": [
    [
      {
        "entity_type": "NAME",
        "start": 6,
        "end": 14,
        "score": 0.995
      }
    ]
  ],
  "stats": { "total_tokens": 4, "tps": 5420 }
}

{
  "text": [
      "I live at <ADDRESS>"
  ],
  "entities": [
      [
          {
              "entity_type": "ADDRESS",
              "output_start": 10,
              "output_end": 19
          }
      ]
  ],
  "stats": {"total_tokens": 13, "tps": 4550.87}
}

{
  "text": [
      "My Nino is #########"
  ],
  "entities": [
      [
          {
              "entity_type": "GOV_SERVICES_ID",
              "output_start": 11,
              "output_end": 20
          }
      ]
  ],
  "stats": {"total_tokens": 11, "tps": 4282.22}
}

{
  "text": [
      "Medicare no. 041827ef234aeddd34e45a4eb9ef23240fdd7072824a372bb0953377557c6d9c"
  ],
  "entities": [
      [
          {
              "entity_type": "HEALTHCARE_ID",
              "output_start": 13,
              "output_end": 77
          }
      ]
  ],
  "stats": {"total_tokens": 10, "tps": 4640.52}
}

For the full range of options, including how to configure entity types via YAML, please visit the documentation

Ready to put your sensitive data to work?

Deploy enterprise-grade PII detection and anonymization on your own infrastructure. Contact us to get started.