Neural style transfer is the concept of blending image's content with a different image's style. This technique helps us generate images that are the same as the base image but with the style of the image. Due to this technique's capacity to produce eye-catching and beautiful visuals, it has experienced substantial growth in popularity. NST, which enables the fusion of content from one image with the creative style of another, has become a fascinating approach in the fields of computer vision and artistic expression. This paper aims to examine how different neural network models and optimization techniques affect the effectiveness and attractiveness of neural style transfer. We specifically compare Adam, Adamax, SGD, RMS Prop, AdamW, Adadelta, Adagrad, and Nadam optimizers with VGG19, ResNet50, ResNet152V2, and Xception.