Imalwayshere/Open-Detector

BERT-based AI-generated academic text detection model

50
/ 100
Established

Built on BERT-base-uncased with binary classification, the model achieves 99.57% accuracy on ~1.4M academic text samples while maintaining a 0.82% false positive rate and 99.94% recall. Weights are distributed via Hugging Face and integrate directly with the `transformers` library, supporting inference through standard PyTorch workflows with 512-token context windows. Beyond stylometric detection, the project roadmap includes fact-checking, citation verification, and hallucination detection to focus on content authenticity rather than writing style alone.

220 stars.

No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 13 / 25
Community 21 / 25

How are scores calculated?

Stars

220

Forks

43

Language

Python

License

MIT

Last pushed

Nov 01, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/Imalwayshere/Open-Detector"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.