So, X removed the mixed text and image layout feature from the post editor, forcing me to use only the "article" tag for adding pictures to my posts. Every article needs an illustration, both for showing off and creating atmosphere, right? So I fed the same prompt to four AI models to generate images: 1 Grok, 2 ChatGPT, 3 Doubao, and 4 Jimeng. Here are the results (Figures 1-4). Clearly, in terms of understanding, color matching, rendering, and visual communication, "Jimeng" stands out from the crowd! I have to say, ByteDance really has something special. Grok… well, let's just say it leans more towards artificial stupidity than artificial intelligence. Is it because it can't understand my Chinese writing? 🤣🤣🤣