---
title: "Transform images from an instruction, letting the server pick the best model"
method: POST
path: "/v2/tool/image-to-image"
tags: ["auto-model"]
---

# Transform images from an instruction, letting the server pick the best model

`POST /v2/tool/image-to-image`

Transform one or more source images by describing the change in plain
language, without choosing a model. The server selects the model best
suited to the request and may route different requests to different
models. Supply the sources either as `image_asset_identifiers`
references (images already stored with Ideogram) or as raw `images`
bytes (multipart requests only) — callers are never required to upload
assets first. If both are supplied, the references win and the bytes
are ignored.

By default the request blocks until the images are ready and returns
them in `data`. Set `async` to true to return immediately after the
request is accepted, then poll for completion and results with
`GET /v1/generations/{generation_id}` using the returned
`generation_id`.

## Request body

- ToolImageToImageRequest — Supply the source images either as `image_asset_identifiers` references or (multipart requests only) as raw `images` bytes. At least one source is required; if both forms are given, the references are used and the bytes are ignored.
  - `prompt` string, required — The instruction describing how to transform the source images.
  - `image_asset_identifiers` AssetIdentifier[] — Existing upload or generated image assets to transform, by reference. Takes priority over `images` if both are supplied.
    - `asset_type` 'ASSET' | 'CANVAS_ASSET' | 'LAYERED_ASSET' | 'RESPONSE' | 'UPLOAD', required
    - `asset_id` string, required
  - `images` string[] — The source images to transform (max 10, max size 25MB per image), as raw bytes; only JPEG, PNG, and WEBP formats are supported. Multipart requests only; ignored if `image_asset_identifiers` is also supplied.
  - `aspect_ratio` 'AUTO' | '1x4' | '1x3' | '1x2' | '9x16' | '10x16' | '2x3' | '3x4' | '4x5' | '1x1' | '5x4' | '4x3' | '3x2' | '16x10' | '16x9' | '2x1' | '3x1' | '4x1' — The aspect ratio for an Ideogram 4.0 magic prompt. `AUTO` lets the model select the most suitable ratio from the prompt; any other value pins the ratio. The non-AUTO values are the buckets the 4.0 model supports.
  - `resolution_tier` '1K' | '2K' | '4K' — The output resolution tier. Influences which model serves the request; not every model offers every tier. When omitted the server uses the selected model's default tier.
  - `magic_prompt` 'AUTO' | 'ON' | 'OFF' — Determine if MagicPrompt should be used in generating the request or not.
  - `num_images` integer — The number of images to generate.
  - `seed` integer — Random seed. Set for reproducible generation.
  - `private` boolean — Whether the generated images should be kept private. When omitted, the default follows the caller's plan; some plans always generate privately.
  - `target_collection_id` string — A collection you can write to, by its URL-safe base64 collection id. The output images are added to it when the request completes.
  - `async` boolean — When false (the default), the request blocks until the images are ready and returns them in `data`. When true, the request returns as soon as it is accepted; poll for completion and results with `GET /v1/generations/{generation_id}` using the returned `generation_id`.

## Response `200`

The transformed images (synchronous requests), or an acknowledgement to poll with `GET /v1/generations/{generation_id}` (`async` requests).

- ToolImageToImageResponse — Response returned by `POST /v2/tool/image-to-image`. Synchronous requests (the default) include the transformed images in `data`. Requests with `async` set to true omit `data`; poll for completion and results with `GET /v1/generations/{generation_id}` using the returned `generation_id`. The seed reports the value the request resolved to when the caller left it unset.
  - `generation_id` string, required — URL-safe base64 ID of the accepted generation. Accepted by the `GET /v1/generations/{generation_id}` polling endpoint.
  - `data` GeneratedImageObject[] — The transformed images, in generation order. Present only for synchronous requests (`async` omitted or false).
    - `url` string, uri, nullable — The direct link to the generated image. Empty when the image did not pass safety checks.
    - `prompt` string, required — The final prompt the image was generated from.
    - `resolution` string, required — The resolution of the generated image, formatted as "WIDTHxHEIGHT".
    - `is_image_safe` boolean, required — Whether the image passed safety checks. If false, `url` is empty.
    - `seed` integer, required — Random seed. Set for reproducible generation.
  - `seed` integer, required — Random seed. Set for reproducible generation.

## Other responses

- `400` — Invalid input provided.
- `401` — Unauthorized.
- `402` — Insufficient credits or quota.
- `404` — A referenced source asset was not found.
- `422` — The prompt did not pass safety checks.
- `429` — Too many requests.
- `500` — Internal server error.
- `503` — The endpoint is temporarily unavailable.

---

[API](https://skmtc.dev/ideogram/apis/ideogram-openapi-3-0.md) · [All operations](https://skmtc.dev/ideogram/apis/ideogram-openapi-3-0/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/ideogram/ideogram-openapi-3-0/revisions/3eb3216eed39/schema)