Image input for GPT-4 (and related docs)

API
1

Hi, from my understanding the image input for GPT-4 and multimodality is not available yet (?).
Is there any preview of the documentation in terms of its capability and how to structure API calls, in order to start learning more about it while awaiting its release?

3 Likes
2

Anyone else also got the same issue here?

{
“object”: “chat.completion”,
“created”: 1679998443,
“model”: “gpt-4-0314”,
“usage”: {
“prompt_tokens”: 53,
“completion_tokens”: 47,
“total_tokens”: 100
},
“choices”: [
{
“message”: {
“role”: “assistant”,
“content”: “I’m sorry, but as an AI language model, I cannot currently view or analyze image files directly. Please provide a description of the issue you’re experiencing with the image or the website, and I’ll do my best to help.”
},
“finish_reason”: “stop”,
“index”: 0
}
]
}

1 Like