跳转到主要内容

​
Vision via Chat

POST /v1/chat/completions
As of March 23, 2026, Crazyrouter production has verified:
  • gpt-5.4 can read images through the OpenAI-style image_url input shape
  • a data:image/png;base64,... data URL succeeds and returns a text answer
  • the current message.content is plain text, not a multipart content array

​
Verified Minimal Request

curl https://crazyrouter.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-5.4",
    "messages": [
      {
        "role": "user",
        "content": [
          {"type": "text", "text": "What color is this image?"},
          {
            "type": "image_url",
            "image_url": {
              "url": "data:image/png;base64,iVBORw0KGgoAAA..."
            }
          }
        ]
      }
    ],
    "max_tokens": 100
  }'
In the 2026-03-23 production check, gpt-5.4 returned: 
{
  "model": "gpt-5.4",
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Red."
      }
    }
  ]
}

​
Current Recommendation

  • Use gpt-5.4 first for new vision-understanding integrations
  • This page only keeps the data-URL path that was revalidated successfully
  • Remote URLs, multi-image inputs, and detail handling were not rechecked item by item in this pass, so revalidate them against production before relying on them