Skip to main content
POST
/
v1
/
responses
o4-mini-deep-research (image analysis)
curl --request POST \
  --url https://api.example.com/v1/responses
OpenAI’s official format(response) for the image analysis API.
curl --location 'https://gptproto.com/v1/responses' \
--header 'Authorization: sk-*****' \
--header 'Content-Type: application/json' \
--data '{
    "model": "o4-mini-deep-research",
    "input": [
        {
            "role": "user",
            "content": [
                {
                    "type": "input_text",
                    "text": "What is in this image?"
                },
                {
                    "type": "input_image",
                    "image_url": "data:image/jpeg;base64,${base64Image}"
                }
            ]
        }
    ]
}'
{
  "id": "resp-abc123",
  "object": "response",
  "created": 1699896916,
  "model": "o4-mini-deep-research",
  "output": "The image shows a wooden boardwalk path extending through a lush green grassland. The boardwalk appears to lead toward a distant tree line under a bright blue sky with some clouds. The grass on either side of the boardwalk is vibrant green, suggesting it might be spring or summer. The scene has a peaceful, natural atmosphere with good visibility and sunny weather conditions.",
  "usage": {
    "prompt_tokens": 1250,
    "completion_tokens": 89,
    "total_tokens": 1339
  }
}

Parameters

ParameterTypeRequiredDefaultDescription
modelstring✅ Yeso4-mini-deep-researchThe model to use for the request. Must be a vision-enabled model.
inputarray✅ Yes-Array of message objects with role and content. Each message contains: - role: “user” or “assistant” - content: Array of content objects supporting the following types: - input_text: Text prompt with text field - input_image: Image input with image_url field, supports: - Base64 encoded images: data:image/jpeg;base64,{base64_string} - Supported formats: JPG, PNG Example with base64 encoded image: json [ { "role": "user", "content": [ { "type": "input_text", "text": "What is in this image?" }, { "type": "input_image", "image_url": "data:image/jpeg;base64,${base64Image}" } ] } ]
streamboolean❌ NofalseWhether to stream the response