OpenMed / token-classification updated 8 months ago

OpenMed-PII-SuperClinical-Small-44M-v1

OpenMed-PII-SuperClinical-Small-44M-v1 is a transformer-based token classification model fine-tuned for Personally Identifiable Information (PII) detection in text. This model identifies and classifies 54 types of sensitive information including names, addresses, SSNs, medical record numbers, and more.

Params
140 M
Context
512
Downloads 30d
247 K
Likes
11
Commercial use: allowed apache-2.0 Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

tracking started — chart appears after 7 days of snapshots (5 recorded)
247 K downloads in the last 30 days

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
model.safetensors f32 0.6 GB 1.1 GB ✅ Runs comfortably
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture
DebertaV2ForTokenClassification
Parameters
140 M
Tensor type
F32
Context length
512
Vocabulary
128,100
Layers / heads
6 / 12
Licence
apache-2.0
First seen on the Hub
2026-01-13
Training datasets
nvidia/Nemotron-PII
nvidia/Nemotron-PII (test_strat) (reported)
0.9529
Added to our catalog
2026-09-16
Compare with any token-classification model