why model “gpt-4-vision-preview” can not process images input? It replys: “I’m sorry, but as a language model AI, I do not have the capability to view images or videos.” But I do set the model parameter as “gpt-4-vision-preview” in the code.
Hi and welcome to the Developer Forum!
There are typically two reasons for issues with this problem.
First, check the ‘max_tokens’ property. It must be adequate for both your prompt and the image. Refer to the OpenAI documentation for guidance on calculating tokens (https://platform.openai.com/docs/guides/vision/calculating-costs).
Second, the structure of your request could be the issue (It happened to me). Errors often occur with incorrect request formats. I recommend reviewing the request structure guidelines in the OpenAI documentation for a better understanding (https://platform.openai.com/docs/guides/vision/quick-start).