> ## Documentation Index
> Fetch the complete documentation index at: https://docs.compose.market/llms.txt
> Use this file to discover all available pages before exploring further.

# Vision

> Use image inputs with chat and Responses-style text models.

Vision means image input on a text route. Pick a model whose catalog row includes `image` in `input` and `text` in `output`.

## Chat completions

```json theme={null}
{
  "model": "gpt-5.5",
  "messages": [
    {
      "role": "user",
      "content": [
        { "type": "text", "text": "What changed in this screenshot?" },
        {
          "type": "image_url",
          "image_url": {
            "url": "data:image/png;base64,...",
            "detail": "auto"
          }
        }
      ]
    }
  ]
}
```

## Legacy attachment fields

Native chat routes also normalize legacy top-level fields:

| Field       | Meaning                                      |
| ----------- | -------------------------------------------- |
| `image_url` | Adds an image part to the last user message. |
| `audio_url` | Adds an audio part to the last user message. |
| `video_url` | Adds a video part to the last user message.  |

Use content parts for new integrations. They are clearer and closer to OpenAI-compatible client behavior.

## Related

* [Text](/inference/modalities/text)
* [Images](/inference/modalities/image)