Generate Image
generate_imageCreate images with HalalGen (the prompt is checked against HalalGen's content rules first: people and animals get a smooth face with no features). 'Auto' uses GPTImage25Flare on paid plans and CF_Flux2Klein on the free plan. Free plan: CF_Flux2Klein and CF_Flux2Dev only. Returns the image URL; share it with the user. To change one of the user's own pictures, use edit_image instead.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"prompt": "<prompt>"
},
"name": "generate_image"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | Image model. Call list_models for costs. | Auto |
| prompt | Yes | What to create. Describe the scene, lighting, materials and mood. Avoid faces. Looks from HalalGen's templates can be asked for here too, for example one object seen from several camera angles, a faceless cartoon-style character, or exercise steps shown with silhouettes. | |
| quality | No | Used by GPT Image 2.6 and 2.5 only: hd and ultra cost more (list_models shows each price). Flux Ultra and Flux 2.0 Pro are always ultra; other models are standard. | standard |
| is_private | No | Keep the image out of the public gallery. Needs a paid plan. | |
| num_images | No | How many images. The free plan makes one; paid plans up to 4. | |
| aspect_ratio | No | Image shape. | 16:9 |
| movement_type | No | Optional camera framing hint added to the prompt. | |
| cinematic_style | No | Optional style preset added to the prompt. |