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

# Generate an image

> Generate a still and reuse it as input to another model.

Text to image is a single request to an image model. Most image models
expose several operations, usually generate and edit, so the request must
name one. Omitting `operation` returns
`error.code: "run.operation_required"` with the valid names in the
message.

The example below uses GPT Image 2, which takes a quality tier and a
resolution alongside the operation name.

```bash theme={null}
curl -X POST https://api.protoface.com/v1/run/openai/gpt-image-2 \
  -H "Authorization: Bearer $PROTOFACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "operation": "image.generate",
    "prompt": "A lighthouse in a storm",
    "quality": "high",
    "resolution": "2k"
  }'
```

The file lands on `image.url` when the run completes, and the `asset_id`
on `outputs` feeds straight into another run, as in
[Animate an image](/guides/image-to-video).

```bash theme={null}
curl https://api.protoface.com/v1/runs/run_... \
  -H "Authorization: Bearer $PROTOFACE_API_KEY"
```

```json theme={null}
{
  "id": "run_...",
  "status": "completed",
  "image": {
    "url": "https://...",
    "content_type": "image/png",
    "file_name": "output.png"
  },
  "outputs": [
    {
      "role": "output",
      "modality": "image",
      "type": "asset",
      "asset_id": "asset_...",
      "file": { "url": "https://...", "content_type": "image/png" }
    }
  ]
}
```