← All models
High-fidelity image editing model with state-of-the-art controllability. Combines JSON + Mask + Image for precise, fine-grained edits ideal for production and enterprise workflows. Trained on licensed data - safe for commercial use.
images
Details
- Accepts
- image
Pricing
- Price
- 5 cr / image
Prices in credits (1 credit = $0.01).
Data schema
Input
| Field | Type | Description |
|---|---|---|
| vgl | — | A pre-built VGL document, edited from verbatim instead of building one from `instruction`. Send this or `instruction`, not both. |
| seed | integer | Random seed for reproducibility. |
| new_vgl | — | The VGL of the image as it should end up. Requires `original_vgl`. |
| mask_url | — | Mask image (file or URL). Optional |
| image_url | — | Reference image (file or URL). |
| steps_num | integer | Number of inference steps. |
| sync_mode | boolean | If true, returns the image directly in the response (increases latency). |
| instruction | — | Instruction for image editing. |
| original_vgl | — | The VGL of the image as it is now. With `new_vgl`, edits by the difference between the two. |
| guidance_scale | — | Guidance scale for text. |
| negative_prompt | string | Negative prompt for image generation. |
Output
| Field | Type | Description |
|---|---|---|
| vgl | object | The VGL document the image was generated from. |
| image | — | Generated image. |
| images | array | Generated images. |
| structured_instruction | object | Deprecated, use `vgl`. The same document under the old name. |