AI Model Security Tool
3.5k 2026-04-30
sentient-agi/OML-1.0-Fingerprinting
A framework for embedding secret cryptographic fingerprints into Large Language Models (LLMs) via fine-tuning to verify ownership and prevent unauthorized use.
Core Features
Generate AI-native cryptographic fingerprints for LLMs.
Embed fingerprints into LLMs through fine-tuning.
Verify LLM ownership and detect unauthorized usage.
Enable users to verify the exact LLM they are interacting with.
Leverages DeepSpeed for scalable fine-tuning.
Quick Start
pip install -r requirements.txtDetailed Introduction
OML 1.0: Fingerprinting provides essential tooling for safeguarding Large Language Models. It introduces an AI-native cryptographic primitive – a unique query-response pair – embedded into LLMs during fine-tuning. This process creates a secret signature, allowing model owners to unequivocally identify their models, detect unauthorized usage, and protect their intellectual property. Furthermore, it empowers legitimate users to verify the authenticity of the LLMs they interact with, fostering trust and accountability in AI deployment.