I trained a CNN model and saved it as a .keras file. Now I want other people to use it for making predictions. I am planning on deploying it using a flask server and package the whole thing in an exe. The problem is When I do a .summary() after I load it back, I am able to see the entire model architecture. I was also able to see the values of the hyperparameters that I used when I trained the model.

What is the best way to serve the model just for inference ? (I want the model specifics to be hidden) and how should the data be passed into whatever I am saving the model/weights as for predictions?
I use keras 2.10.0 and tf 2.10.1.

The most straightforward approach in the keras docs was to use .export() but it seems that it is available only from keras 3.