300 tokens for a 512x512 image at "detail: low"

API
, ,
1

I use sample from documentation(https://platform.openai.com/docs/guides/vision):

from openai import OpenAI
client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {
                    "type": "image_url",
                    "image_url": {
                        "url": "https://my.site....",
                        "detail": "low",
                    },
                },
            ],
        }
    ],
    max_tokens=10,
)

print(response.choices[0].message.content)

i use this image(512x512)

A colorful fruit platter features sliced pineapple, pears, oranges, grapes, strawberries, and a garnishing of mint. (Captioned by AI)
A colorful fruit platter features sliced pineapple, pears, oranges, grapes, strawberries, and a garnishing of mint. (Captioned by AI)512×512 106 KB

BUT!!!
In the response I see:

prompt_tokens: 303
completion_tokens: 10
total_tokens: 313

Why 303 tokens?

On the same page of the documentation it says 85 tokens for a 512x512 image and “low detail”
https://platform.openai.com/docs/guides/vision

Why 303 tokens?

2 Likes