The choice of a model usually depend on the use case.In general the most widely used and highly regarded models for imageclassification are:
ResNet is a deep residual network that is developed by Microsoft.It was designed to alleviate the vanishing gradient problem that plagues many deep neural networks
ResNet
Inception, also known as GoogleNet, is a CNN developed by Google for the ImageNet Large Scale Visual Recognition Challenge
Inception
The VGG network, developed by the a Group at the University of Oxford.It is known for its simplicity and the fact that it uses small, 3x3 filters throughout the network
VGG
MobileNet is a light-weight CNN designed for efficient inference on mobile and embedded devices. It is particularly well-suited for tasks with limited computational resources.
MobileNet
NASNet is a CNN that was developed using neural architecture search.NASNet is known for its high accuracy and efficiency, but it also has a large number of parameters.