Confronto tra mappe alimentari di Guangzhou tramite intelligenza artificiale

This is a comparison image with a black background and a faint gray grid, designed in the style of a social media test result poster. At the top are two side-by-side bold sans-serif titles: “GPT image 2” above the left image and “Nano Banana” above the right image. Below are two vertically arranged illustration posters with a small gap between them. Both posters depict the same concept: a hand-drawn-style Guangzhou food walking map, but presented in two different visual styles. The left poster is more detailed, textured, dense in content, and features refined illustrations; the right poster is more concise, flat, and intuitive. The left poster uses a warm-toned cream paper background, presenting a refined retro hand-drawn travel map style with delicate lines, watercolor texture, decorated with flowers and lychees, and depicting miniature buildings, roads, rivers, and numerous food illustrations. The large Chinese title at the top of the left poster reads “Guangzhou Foodie Walking Map.” The map covers approximately 12 food spots located throughout the city, each with a small food illustration, Chinese label, and a short handwritten-style annotation. Visible foods include skewers, roast meats, noodles, clay pot rice, rice noodle rolls, dim sum, soups, and sweet bowls, with landmarks like the Canton Tower and the Pearl River also integrated. The right poster also features the Chinese title “Guangzhou Foodie Walking Map,” showcasing the same concept but in a more concise infographic map style, simplifying roads, icons, flower and lychee border decorations, food labels, and depicting rivers, the Canton Tower, and approximately 12 labeled food illustrations. At the bottom of the entire comparison image is a rounded-corner black text box with a white outline. A bright yellow circular label with bold black Chinese characters “Prompt” is attached to the upper left corner. The black panel contains a white Chinese text describing the image generation prompt in multiple lines of paragraphs. The overall composition is centered, with high contrast, presenting a refined AI image comparison post suitable for Chinese social media, featuring clear layout, poster-like symmetry, and a concise test benchmark aesthetic.