I've created a benchmark of 10 text-to-image tasks and matched 10 pictures for each task, which are pictures that you can easily find on Google. It is extremely easy to check that for every task, all the current AIGC platforms are totally incapable of creating pictures that can fulfill them, so this is a benchmark that can be used to check the intelligence of future AIGC frameworks.
I greatly respect how image benchmarks like MNIST and Imagenet have contributed to AI advancements, so this is my attempt at developing a benchmark. Please leave feedback.