Can we use gradient desent method in maximum entropy model?

I see a lot of implementations use GIS or IIS to train the maximum entropy model. Can we use gradient desent method? If we can use it, why most tutorial directly tell GIS or IIS methos, but do not show the simple gradient desent method to train maximum entropy model? As we know, softmax regression is equivalent to the maxent model, but I never heard GIS or IIS in softmax. Is there a toy code use simple gradient desent method to train maxent model?