Skip to main content
Models targeting DEEPX accelerators via ONNX Runtime can be verified before creating an inference session with the secure loader: 
import onnxruntime as ort
from thistle_secure_loader import secure_load

def deepx_loader(path: str):
    return ort.InferenceSession(path)

session = secure_load("model.onnx", deepx_loader)
print("DEEPX model verified.")
The secure_load call verifies the model file’s signature using tuc before passing it to ort.InferenceSession. If the signature check fails, a ModelVerificationError is raised and the model is never loaded.