Vision API: Where to insert the "High / Low" Detail Parameter?

Vision API: Where to insert the “High / Low” Detail Parameter?

The documentation talks from a Low/High Detail Parameter.
But there is no Sample how or where it needs to be specified,.

## Low or high fidelity image understanding

By controlling the detail parameter, which has two options, low or high, you have control over how the model processes the image and generates its textual understanding.


I have exactly the same problem ! If anyone has a solution, we’re interested! Especially since the costs seem drastically lower with the “low” mode, we should see if this does not crush the performance too much.

It’s under the “image_url” key.
In the same scope as “url” key.

    payload = {
        "model": "gpt-4-vision-preview",
        "messages": [
            "role": "user",
            "content": [
                "type": "text",
                "text": f"{prompt}"
                "type": "image_url",
                "image_url": {
                    "url": f"data:image/jpeg;base64,{base64_image}",
                    "detail": "low"
        "max_tokens": 300

Thanks for sharing.
Can you tell me something about the minimum resolution for the uploaded images?
To me looks that small images can not be interpreted.