Ideogram 4.5 Edit API

ideogram/v4.5/edit

image_urls · mask_url · reference_image_urls

Ideogram 4.5 Edit turns one source image and a written instruction into a revised image, with optional masks, up to four reference images and high edit precision. It keeps the areas you did not ask to change, so repeated edits stay close to the original.

Get API Key

Input

PromptRequired169/10000

Image UrlsRequired

Drag and drop files, paste from clipboard, or provide a URL. Exactly one source image (public HTTP(S) URL).

1/1 files added

Mask UrlOptional

Drag and drop files, paste from clipboard, or provide a URL. Optional mask, same size as the source: black = edit, white = keep. Requires size auto.

Reference Image UrlsOptional

Drag and drop files, paste from clipboard, or provide a URL. Up to 4 reference images, or 3 when a mask is set. They guide the result and are not edited.

Edit PrecisionOptional

QualityOptional

SizeOptional

NOptional

OutputReady

Estimated cost: 12 credits · $0.0600

Credits are reserved when the task is submitted and refunded in full if it fails.

Continue using

Ideogram 4.5 Text to Image FLUX 3 Klein Edit Nano Banana 2.1 Edit

Examples

Replace the masked sky with a clear late-afternoon sky with a few soft clouds. Keep the house, garden and shadows unchanged.

Place the ceramic mug from the source image on a linen tablecloth in the morning-light kitchen shown in the reference image. Keep the mug shape, glaze colour and handle exactly as in the source.

REST API

Quick Start

This flow submits one valid Ideogram 4.5 Edit request, captures its task_id, and reads result file URLs after the asynchronous task ends.

Connect to the Vidgo API

Create an API key, keep it on the server, and send it as Authorization: Bearer VIDGO_API_KEY.

Endpoint: POSThttps://api.vidgo.ai/api/generate/submit
Authentication: Authorization: Bearer VIDGO_API_KEY

Submit one task

Send the current model ID and an input object to the unified submit endpoint. input may only contain the fields listed for this endpoint; any other field returns 400.

Submit request

curl --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "ideogram/v4.5/edit",
  "input": {
    "prompt": "Change the chalkboard sign text to read exactly \"FRESH BREAD DAILY\" in white chalk lettering. Keep the storefront, lighting, wood grain and every other detail unchanged.",
    "edit_precision": "high",
    "quality": "medium",
    "size": "auto",
    "n": 1,
    "image_urls": [
      "https://cdn.vidgo.ai/apis/models/ideogram/v4.5/edit/v1/01/input-01-source-placeholder.png"
    ]
  }
}'

Wait for the result

Read data.task_id and poll only while status is not_started or running. Stop on both finished and failed.

Track status

GET https://api.vidgo.ai/api/generate/status/{task_id}

Poll every 2–5 seconds and back off on 429/5xx. Or provide top-level callback_url to receive a terminal-state callback.

not_startedrunningfinishedfailed

Submit response

{
  "code": 200,
  "data": {
    "task_id": "task-example",
    "status": "not_started",
    "created_time": "2026-11-03T08:00:00"
  }
}

Finished response

{
  "code": 200,
  "data": {
    "task_id": "task-example",
    "status": "finished",
    "created_time": "2026-11-03T08:00:00",
    "progress": 100,
    "files": [
      {
        "file_url": "https://example.com/result.png",
        "file_type": "image"
      }
    ]
  }
}

Complete runnable polling example

Expand for the same payload end to end, including business-code, missing-task-ID, failed-state, unexpected-state, and network-error handling.

Complete runnable polling example

set -euo pipefail
: "${VIDGO_API_KEY:?Set VIDGO_API_KEY in your environment}"

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "ideogram/v4.5/edit",
  "input": {
    "prompt": "Change the chalkboard sign text to read exactly \"FRESH BREAD DAILY\" in white chalk lettering. Keep the storefront, lighting, wood grain and every other detail unchanged.",
    "edit_precision": "high",
    "quality": "medium",
    "size": "auto",
    "n": 1,
    "image_urls": [
      "https://cdn.vidgo.ai/apis/models/ideogram/v4.5/edit/v1/01/input-01-source-placeholder.png"
    ]
  }
}
JSON
)

SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

TASK_ID=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.data.task_id // .task_id // empty')
if [ -z "$TASK_ID" ]; then
  printf 'Submit response did not include task_id:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi

while true; do
  STATUS_RESPONSE=$(curl --silent --show-error --fail-with-body \
    --url "https://api.vidgo.ai/api/generate/status/$TASK_ID" \
    --header "Authorization: Bearer $VIDGO_API_KEY")
  STATUS=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.data.status // .status // empty')

  case "$STATUS" in
    finished)
      printf '%s' "$STATUS_RESPONSE" | jq -r '(.data.files // .files // [])[]?.file_url'
      break
      ;;
    failed)
      printf '%s' "$STATUS_RESPONSE" | jq -r '.data.error_message // .error_message // "Generation failed"' >&2
      exit 1
      ;;
    not_started|running)
      sleep 2
      ;;
    *)
      printf 'Unexpected task status: %s
' "$STATUS" >&2
      exit 1
      ;;
  esac
done

Request Parameters

This table lists only public fields inside input. See Quick Start for top-level model and optional callback_url usage.

Field	Type	Required	Default	Description
input.prompt	string	Yes	–	Edit instruction, non-blank, up to 10,000 characters.
input.image_urls	string[]	Yes	–	Exactly one source image as an array with one public HTTP(S) URL or image data URI.
input.mask_url	string	No	–	Optional mask the same size as the source: black = edit, white = keep. Requires size auto.
input.reference_image_urls	string[]	No	–	0–4 reference images (0–3 when mask_url is set). They guide the result and are not edited.
input.edit_precision	string	No	regular	regular or high. high requires size auto.
input.quality	string	No	medium	very_low, low, medium or high. Sets the price per image.
input.n	integer	No	1	Number of images, integer 1–8.
input.size	string	No	auto	auto keeps the source size. Otherwise a preset (1:1, 1:1 HD, 4:3, 3:4, 16:9, 9:16) or WIDTHxHEIGHT with sides ≥256 and multiples of 32, at most 4,194,304 pixels and 6:1. Must be auto with a mask or high precision.

Response Fields

The submit response identifies the task. The status response supplies result files or failure details at a terminal state.

Field	Type	Description
code	integer	Business response code; 0 or 200 indicates success.
message	string	Optional human-readable response message.
data.task_id	string	Unique task ID used for status queries after submission.
data.status	string	not_started, running, finished, or failed.
data.created_time	string	Task creation timestamp.
data.progress	integer	Progress from 0–100 in status responses.
data.files[]	array	Generated files in returned order when finished.
data.files[].file_url	string	Direct URL of the generated file (.png).
data.files[].file_type	string	Result files are identified as image.
data.error_message	string \| null	Readable failure reason for failed tasks.

Task Lifecycle

Only the first two states may be polled again. finished and failed are terminal.

not_started
Accepted and queued; wait before checking again.
running
Generation is in progress; continue polling.
finished
Successful terminal state; validate files and read file URLs.
failed
Failure terminal state; read error_message and stop. Reserved credits are refunded in full.

Polling and Errors

AuthenticationKeep a valid VIDGO_API_KEY on the server and send requests with Bearer authentication.
Validation400 indicates an invalid model ID, field, enum, length, count or size for this endpoint, including fields that belong to another endpoint.
Polling intervalStart at 2–5 seconds, use exponential backoff for 429/5xx, and enforce an overall timeout.
Terminal statesRead file URLs from files after finished, or error_message after failed. Both states end polling.
Callback optionProvide a public top-level callback_url to receive the task object after finished or failed.
Network errorsDistinguish request timeout, network failure, and task failure. Retry the same query only when safe.

Model Specifications

Specification	Value	Details
Model	ideogram/v4.5/edit	Ideogram 4.5 Edit
Input	prompt + image_urls (1)	Optional mask_url and 0–4 reference_image_urls
Output	PNG image × n (1–8)	Raster image URLs in files[]
quality	very_low / low / medium / high	1.6 / 6 / 12 / 44 credits per image
size	auto / presets / WIDTHxHEIGHT	Default auto keeps the source size

View the Ideogram 4.5 Edit API docs

Ideogram 4.5 Edit

Ideogram 4.5 Edit is the editing endpoint of Ideogram 4.5, a model Ideogram positions as its most precise editor. It takes exactly one source image, an instruction of up to 10,000 characters, an optional black-and-white mask and up to four reference images that guide style or content without being edited. Four quality tiers start at 1.6 credits, and the default size auto keeps the source dimensions.

Why Choose Ideogram 4.5 Edit?

Precise, instruction-based editsDescribe the change in plain language; Ideogram says 4.5 is built to avoid pixel shifts and colour drift across repeated edits.
Masks for targeted regionsSupply mask_url to limit the edit: black marks the area to change, white the area to keep. Masked edits run at size auto.
Reference images that guide, not changeAdd up to four reference_image_urls (three when a mask is used) to steer style, objects or setting while only the source is edited.
High edit precisionSet edit_precision to high to restore untouched pixels from the source so only the requested change differs.
Four tiers from 1.6 creditsvery_low, low, medium and high cost 1.6, 6, 12 and 44 credits per image; masks, references and size never change the price.

Parameters

Parameter	Requirement	Description
prompt	Required	Edit instruction, non-blank, up to 10,000 characters.
image_urls	Required	Exactly one source image as an array with one public HTTP(S) URL or image data URI.
mask_url	Optional	Optional mask the same size as the source: black = edit, white = keep. Requires size auto.
reference_image_urls	Optional	0–4 reference images (0–3 when mask_url is set). They guide the result and are not edited.
edit_precision	Optional	regular or high. high requires size auto. Default`regularhigh`
quality	Optional	very_low, low, medium or high. Sets the price per image. Default`mediumvery_lowlowhigh`
n	Optional	Number of images, integer 1–8. Default`1`
size	Optional	auto keeps the source size. Otherwise a preset (1:1, 1:1 HD, 4:3, 3:4, 16:9, 9:16) or WIDTHxHEIGHT with sides ≥256 and multiples of 32, at most 4,194,304 pixels and 6:1. Must be auto with a mask or high precision. Default`auto`

How to Use

Prepare the sourceUse one public image URL. For a masked edit, also prepare a mask with the same width and height, black where the change goes.
Write the instructionSay what to change and what must stay, and add reference images if a style, product or scene should guide the result.
Pick precision, quality and sizeUse high edit precision for small surgical changes, keep size auto unless you need a new format, and choose a quality tier.
Submit and collectSend model ideogram/v4.5/edit to the submit endpoint, then poll with the task_id or wait for callback_url to read the edited images.

Usage and Pricing

Billed per output image by quality. Total credits = rate × n. 1 credit = $0.005.

Credits are reserved when the task is submitted and refunded in full if it fails. Masks, reference images, edit precision and size do not change the price.

Usage and Pricing
Option	Credits / image	USD / image
very_low	1.6	$0.008
low	6	$0.03
medium	12	$0.06
high	44	$0.22

Best Use Cases

Localising packaging and signageReplace the words on a label, menu board or sign while the material, lighting and perspective stay as photographed.
Product scene swapsKeep the product from the source and use a reference image to move it into a new setting for listings or ads.
Staged interior revisionsMask a sofa, wall or floor and restyle it while the rest of the room is preserved.
In-app edit featuresLet users brush a region in your app, send it as mask_url and return the edited image from one API call.

Pro Tips

Mask convention: the mask must match the source size and contain both black and white; black areas are edited, white areas are kept.
A mask uses one reference slot, so plan for at most three reference images when you send mask_url.
For successive edits, feed the finished output URL back as the next request's image_urls; each step is billed separately.
Use very_low to test an instruction quickly, then rerun the same request at medium or high.

Notes

image_urls must contain exactly one image. Additional pictures go in reference_image_urls.
Sending mask_url or edit_precision high with a size other than auto returns 400.
Custom sizes must use sides of at least 256 px in multiples of 32, at most 4,194,304 pixels in total and an aspect ratio no wider than 6:1.

Related Models

GPT Image 2.6 Flare Text to Image preview

GPT Image 2.6 Flare Text to ImageGPT Image 2.6 Flare Text to Image turns text prompts into 1K–4K images with fast low-quality drafts, legible poster typography and transparent backgrounds. It places headlines, subjects and props where the prompt puts them, so one prompt carries a concept from quick draft to final quality.

GPT Image 2.6 Flare Image EditGPT Image 2.6 Flare Image Edit turns 1–16 reference images and text instructions into a new 1K–4K image, with optional masks, transparent cutouts and adjustable planning effort. It keeps the first image’s aspect ratio by default, and keep-and-change instructions let it recolor or relight a subject while the rest holds.

GPT Image 2.6 Sunburst Text to Image preview

GPT Image 2.6 Sunburst Text to ImageGPT Image 2.6 Sunburst Text to Image turns detailed text prompts into 1K–4K images with fine material texture, crowded multi-element layouts and quality up to max. It keeps long scene descriptions in order, placing subjects, props and light where the prompt puts them.

GPT Image 2.6 Sunburst Image Edit preview

GPT Image 2.6 Sunburst Image EditGPT Image 2.6 Sunburst Image Edit turns 1–16 reference images and text instructions into a refined 1K–4K image, with masks, multi-image composites and quality up to max. Built for stronger consistency across edits, it keeps identity, materials and lighting steady while applying the requested change.

Ideogram Vector 1.0 Text to Vector preview

Ideogram Vector 1.0 Text to VectorIdeogram Vector 1.0 Text to Vector turns a written prompt into SVG artwork for logos, icons and illustrations, with up to five reference images and 17 aspect ratios plus auto. Results are true vector files that scale cleanly from favicon to billboard.

Ideogram Vector 1.0 Image to Vector preview

Ideogram Vector 1.0 Image to VectorIdeogram Vector 1.0 Image to Vector turns one raster image into an SVG file, with a detailed default mode and a simple mode that uses fewer shapes. It follows the shapes and colours of the source so logos, sketches and illustrations become scalable artwork.

FLUX 3 Klein Text to ImageFLUX 3 Klein Text to Image turns prompts into product shots, scenes and illustrations at 1K, 2K or 4K, with seven presets, exact custom sizes and up to four images per call. It keeps per-image cost low so you can iterate widely before rendering finals.

FLUX 3 Klein EditFLUX 3 Klein Edit turns one to ten input images and a written instruction into edited or combined results at 1K, 2K or 4K, priced the same as generation. It keeps the subjects you name from each image while changing backgrounds, outfits, materials or style.

Nano Banana 2.1 Text to ImageNano Banana 2.1 Text to Image turns text prompts into 1K–4K images with text layout rendering and panoramic compositions. It preserves perspective and lighting while eliminating tiling artifacts in panoramic aspect ratios.

Nano Banana 2.1 EditNano Banana 2.1 Edit uses 1–10 reference images for image blending and local edits, with consistent characters and sharp redraws. It preserves character appearance and lighting as elements change, delivering crisp results at up to 4K resolution.

Ideogram 4.5 Text to ImageIdeogram 4.5 Text to Image turns written prompts into posters, wordmarks and text-rich images, with three quality tiers and 1–8 images per request. It keeps headlines and short copy legible while you pick the aspect ratio or one of 36 exact pixel sizes.

Qwen Image 2.1 Text to ImageQwen Image 2.1 Text to Image turns text into native 1K and 2K images with RGBA transparency and Chinese-English typography. It preserves lighting, composition, and aspect ratios while delivering clean transparent layers that need no background removal.

Qwen Image 2.1 EditQwen Image 2.1 Edit turns 1-10 reference images and instructions into precise edits, with multi-image feature fusion and masked inpainting. It preserves facial features and product materials while replacing backgrounds naturally and retaining every pixel outside the mask.

Seedream 5.0 Flash Text to Image preview

Seedream 5.0 Flash Text to ImageSeedream 5.0 Flash Text to Image turns text prompts into high-definition 1K to 2K images, with Chinese and English text layout and 8 aspect ratios. It closely follows complex prompts while providing fine control over lighting and composition.

Seedream 5.0 Flash Image EditSeedream 5.0 Flash Image Edit combines 1 to 10 reference images with a prompt to create new work at up to 2K resolution. It closely preserves faces, clothing and composition from the source images while naturally transforming their style.

GPT Image 2.5 Flare Text to Image preview

GPT Image 2.5 Flare Text to ImageGPT Image 2.5 Flare Text to Image turns text prompts into detailed 1K–4K images, with transparent backgrounds and low latency. It combines fast responses with natural lighting while avoiding structural distortion in large images.

Ideogram 4.5 Edit API FAQ

What is the Ideogram 4.5 Edit API?

Ideogram 4.5 Edit is an Ideogram model for editing an existing image from a written instruction. It revises one source image with an optional mask, up to four reference images and regular or high edit precision, returning one to eight results at four quality tiers. Ideogram designed it to keep untouched areas stable so repeated edits do not accumulate drift. You can call it programmatically or try it from the playground above.

How does the mask work in Ideogram 4.5 Edit?

The mask must have the same width and height as the source and contain both black and white. Black marks the area to edit and white the area to keep. A masked request must use size auto.

How many reference images does Ideogram 4.5 Edit accept?

Up to four in reference_image_urls, or three when mask_url is set. They guide the result and are never edited themselves; the single image to edit goes in image_urls.

What does high edit precision do in Ideogram 4.5 Edit?

edit_precision high restores unchanged pixels from the source so that only the requested change differs. It requires size auto.

How much does Ideogram 4.5 Edit cost?

Per output image: very_low 1.6 credits ($0.008), low 6 ($0.03), medium 12 ($0.06) and high 44 ($0.22). Total = rate × n; masks, references, precision and size do not affect the price.

Can Ideogram 4.5 Edit change the output size?

Yes for regular unmasked edits: use a preset or a custom WIDTHxHEIGHT with sides ≥256 in multiples of 32, up to 4,194,304 pixels and 6:1. The default auto keeps the source size.

How do I chain multi-step edits with Ideogram 4.5 Edit?

Use the finished output URL as image_urls in the next request and describe only the next change. Each step is an independent task billed by quality × n.

Can Ideogram 4.5 Edit replace text inside a photo?

Yes. Quote the new wording in the instruction and say which sign or label it replaces; adding a mask over that area keeps the rest of the photo untouched.