Detect PII and PHI with NVIDIA's GLiNER-PII Model

Melvin Vivas · X post · 2026-04-15 · Open on X

Topics: AI Safety, Security & Guardrails · Level: intermediate

Summary

Melvin shares NVIDIA's GLiNER-PII model on Hugging Face. It finds Personally Identifiable Information (PII) and Protected Health Information (PHI) in text. He also links a Hugging Face Space where you can test it in the browser.

Key points

• PII means Personally Identifiable Information. PHI means Protected Health Information.
• nvidia/gliner-PII is a GLiNER-based named-entity model built to classify PII and PHI.
• You can try it with no setup in the 'PII Guardian' Hugging Face Space.
• Models like this work well as a privacy guardrail: redact sensitive data before you send text to an LLM or store it.

Resources mentioned

• GLiNER-PII · tool · huggingface.co · free
  NVIDIA's GLiNER-based model that finds personal data (PII) in text, so it can be redacted locally.
  Also in: aibackends 0.3.0: Model Caching Speeds Up PII and OCR Inference (Melvin Vivas on X · notes), Running AI Locally: Rebuilding AIBackends as a Python Library with Open Models (Melvin Vivas on X · notes)
• PII Guardian (Hugging Face Space) · website · huggingface.co · free
  A Hugging Face Space where you can try the GLiNER-PII model in your browser.
• Hugging Face · website · x.com · free · recommended by both Bashiri Smith & Melvin Vivas
  Platform for hosting and finding ML models, datasets and papers. The quoted post says LocateAnything was trending there.
  Also in: Using an ML agent to train an open-source TTS model on your voice (Melvin Vivas on X · notes), Deploy Open-Source Models with Hugging Face Inference Endpoints (Melvin Vivas on X · notes), Deploying Qwen3.8 27B on Hugging Face Inference Endpoints (Melvin Vivas on X · notes), Using Hugging Face credits: Jobs, Inference Endpoints and Open Models (Melvin Vivas on X · notes) and 38 more

Try this

• Test nvidia/gliner-PII on sample text in the PII Guardian Space.
• Look for other PII/PHI detection models to compare against it.
• Build a PII/PHI redaction guardrail that cleans user input before it reaches an LLM.