PasteProof PII Detector - ONNX (Browser-Ready)

Quantized ONNX version of pasteproof-pii-detector-v3 for fast browser-side inference.

Performance

Metric	Value
Model size	~147 MB
Inference time	50-100ms (browser)
Accuracy	~97% F1 (minimal loss from quantization)

Usage with Transformers.js

import { pipeline } from '@xenova/transformers';

const detector = await pipeline(
  'token-classification',
  'joneauxedgar/pasteproof-pii-detector-onnx'
);

const results = await detector('const key = "sk_live_abc123";');
console.log(results);

Usage with ONNX Runtime Web

import * as ort from 'onnxruntime-web';

const session = await ort.InferenceSession.create('model.onnx');
// ... tokenize and run inference

Files

model.onnx - Quantized ONNX model
tokenizer.json - Tokenizer vocabulary
tokenizer_config.json - Tokenizer settings
config.json - Model configuration
special_tokens_map.json - Special token mappings

Original Model

See pasteproof-pii-detector-v3 for full details on entity types and training.

Downloads last month: 2