Ideogram AI Ideogram V4.6 Edit API Documentation

Ideogram AI Ideogram V4.6 Edit API Documentation

Playground

Try it on WaveSpeedAI!

Ideogram V4.6 Image Edit supports written editing instructions, reference images, optional masks and precision controls for product visuals, recoloring and creative variations. Available through the WaveSpeed REST API.

Features

Ideogram V4.6 Edit changes an existing image from a written instruction, with optional reference images and a mask. Use it to explore product variations, recoloring and campaign visuals.


Why Choose This?

  • Prompt-based editing — Change objects, colors, materials, text and layout with natural language.
  • Reference images — Up to four supporting images guide subjects, products or style.
  • Masked edits — Use a mask to guide a local edit.
  • High precision — Use edit_precision=high for focused edits. See Notes for output-size behavior.
  • Resolution options — Choose 1k, 2k or 4k; output dimensions depend on the editing mode.

Parameters

ParameterRequiredDescription
imageYesPublic URL of the source image.
promptYesThe edit to make (1–10000 characters).
reference_imagesNoUp to 4 guidance images, or 3 with mask_url.
mask_urlNoOptional mask for a local edit. Use the same dimensions as the source image.
aspect_ratioNo1:1, 4:3, 3:4, 3:2, 2:3, 16:9 or 9:16. For regular edits without a mask, omit it to follow the source aspect ratio. Ignored with a mask or high precision.
resolutionNo1k, 2k or 4k. Default: 1k. For mask-based or high-precision edits, see Notes.
edit_precisionNoregular or high. Default: regular.
qualityNovery_low, low, medium or high. Default: medium. very_low is available up to 2k.

Pricing

Quality1k / 2k4k
very_low$0.008—
low$0.03$0.06
medium$0.06$0.20
high$0.22$0.60

A default edit (1k, medium) costs $0.06 per image. Billing follows the selected quality and resolution tiers. Reference images, the mask and edit_precision do not add a separate fee.


Best Use Cases

  • Product variations — New colors, materials and finishes on the same product.
  • Brand recoloring — Explore campaign color palettes.
  • Selective fixes — Repair or replace one area with a mask.
  • Character variations — Use reference images to guide character appearance in new scenes.

Pro Tips

  • Say what should change and what must stay the same.
  • Use a mask or edit_precision=high for focused edits, then review the result for unintended changes.
  • Keep reference images closely related to the change you want.

Notes

  • very_low quality supports 1k and 2k; choose low, medium or high for 4k.
  • Supply up to four reference images without a mask, or up to three when using mask_url. The source image is separate from these references.
  • With mask_url or edit_precision=high, output dimensions follow the source rather than aspect_ratio. Larger images may be resized; selecting 4k does not enlarge a smaller source to 4K. The selected quality and resolution tiers still determine the price.
  • For very wide or tall source images, select a listed aspect_ratio when using regular editing without a mask.
  • Review generated text, colors, subject details and areas outside the intended edit before using the result.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "1:1",
  "resolution": "1k",
  "edit_precision": "regular",
  "quality": "medium"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/ideogram-ai/ideogram-v4.6/edit" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Retry temporary errors only while polling, never the submit request. Errors, retries and timeouts

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
imagestringYes1 ~ unlimited charactersSource image to edit. Supply a publicly accessible image URL.
promptstringYes1 ~ 10000 charactersDescribe the changes to make to the source image.
reference_imagesarray<string>No-0 ~ 4 items · each item: 1 ~ unlimited charactersOptional guidance images. Up to four without a mask, or three when mask_url is supplied.
mask_urlstringNo-1 ~ unlimited charactersUpload a mask image or provide its URL for a local edit. Use the same dimensions as the source image.
aspect_ratiostringNo-1:1, 4:3, 3:4, 3:2, 2:3, 16:9, 9:16Output aspect ratio for regular edits without a mask. If omitted, it follows the source aspect ratio. Ignored with mask_url or edit_precision=high.
resolutionstringNo1k1k, 2k, 4kSelect a resolution and price tier: 1k, 2k or 4k. With a mask or high precision, dimensions follow the source and larger images may be resized. Selecting 4k does not enlarge a smaller source.
edit_precisionstringNoregularregular, highChoose regular editing or high precision for focused changes. With high precision, dimensions follow the source and larger images may be resized.
qualitystringNomediumvery_low, low, medium, highQuality and price tier. very_low is available up to 2k.

Conditional constraints

  • Not allowed: reference_images has at least 4 items and mask_url is provided.
  • Not allowed: quality is very_low and resolution is 4k.

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.statusstringTask status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses.
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.statusstringStatus: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.