About Ideogram v4.5 Edit
Ideogram v4.5 Edit transforms a source image using text instructions, with optional reference images and masks for targeted changes.
Regular editing supports broad transformations, while high precision restores unchanged pixels to preserve areas outside the edit. Up to four reference images can guide an edit, reduced to three when a mask is supplied. Masks must match the source dimensions: black regions are edited and white regions are preserved, and both regions must be present.
Output preserves the source geometry by default, with downscaling when needed. Only unmasked regular edits can pick a different aspect ratio and resolution (1K or 2K), and Ideogram prices each image by quality tier only. Four quality tiers cover quick iteration through final designs.
Ready to build with Ideogram v4.5 Edit?
Try Ideogram v4.5 Edit in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/images/edit \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "ideogram-v4-5-edit",
"prompt": "A cat sitting on a chair",
"input_image": "https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/images/ox_zoom_out_1926_1076.png",
"edit_precision": "regular",
"quality": "medium",
"aspect_ratio": "auto",
"resolution": "1K",
"sync_mode": false
}'Request parameters
Every field Ideogram v4.5 Edit accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/images/edit| Field | Type | Required | Default | Description |
|---|---|---|---|---|
prompt | string | Yes | — | Describe how to edit the source image, using the optional reference images for guidance. |
input_image | stringuri | Yes | — | Primary source image to edit. |
input_images | array<string> | No | — | Up to four reference images, or three when a mask is supplied, in addition to the primary source image. |
mask_url | stringuri · nullable | No | — | Optional mask matching the source dimensions. Black pixels mark regions to edit; white pixels are preserved. Must contain both regions. When supplied, keep the aspect ratio on auto and use at most three reference images. |
edit_precision | string | No | "regular" | Regular performs generative editing. High uses precise editing with restoration of unchanged pixels.regularhigh |
quality | string | No | "medium" | Image quality tier. Very low costs $0.008, low $0.03, medium $0.06, and high $0.22 per image, independent of image size.very_lowlowmediumhigh |
aspect_ratio | string | No | "auto" | auto1:116:99:164:33:43:22:35:44:516:1010:162:11:2 |
resolution | string | No | "1K" | 1K2K |
image_size | object | No | — | Pixel dimensions sent to fal, looked up from aspect_ratio and resolution in the size table below. Auto passes through unchanged. |
seed | integernullable | No | — | Random seed. Omit to choose automatically. |
sync_mode | boolean | No | false | Return image data directly in the response. |