ChatGPT AI 图像提示词

第 43 / 515 页

ChatGPT image prompts in this library were written for OpenAI's image models, including the GPT Image line. They are production notes, not a pile of quality adjectives. A usable one names who or what is in the frame, where the camera sits, how the light falls, and what the model must not invent. Start here when the picture will actually be generated in ChatGPT and you need a prompt that already survived that model's habits: it likes to tidy faces, spell nonsense on signs, and add a second object you did not ask for. The text on each page is the original prompt. This hub does not translate it or rewrite it into another model's dialect. Use the portrait, product, poster, and photorealistic hubs beside it when you need a narrower job. If a page also carries a Midjourney or Flux label, that is a separate fact about the prompt, not a claim that the same wording transfers unchanged.

更多模型提示词