I got a similar result with ChatGPT; however, it has no issues with written instructions. Interestingly, when I asked it to convert the written instructions into an image, it still gave a result similar to OP’s.
It seems there’s a fundamental incapacity for the model to produce an ordered series of images inside one image. What if you ask to describe each step with a separate image?
I got a similar result with ChatGPT; however, it has no issues with written instructions. Interestingly, when I asked it to convert the written instructions into an image, it still gave a result similar to OP’s.
It seems there’s a fundamental incapacity for the model to produce an ordered series of images inside one image. What if you ask to describe each step with a separate image?