Bytedance Seedream V5.0 Lite Edit API Documentation

Bytedance Seedream V5.0 Lite Edit API Documentation

Playground

Seedream 5.0 Lite Edit by is a state-of-the-art image editing model preserving facial features, lighting, and color tones from reference images. Features high-fidelity editing with professional quality, superior prompt adherence, and up to 4K resolution. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Seedream V5.0 Lite Edit transforms existing images with text instructions. Use reference pictures to guide changes to subjects, backgrounds, styles, and compositions, with custom sizing and JPEG or PNG output.

Why Choose This?

• Reference-guided creation — Use up to 10 images to guide subjects, styles, and compositions.
• Natural language editing — Describe the changes you want in plain text.
• Creative transformations — Explore new backgrounds, outfits, materials, and visual styles.
• Flexible output — Choose custom dimensions and JPEG or PNG output.

Parameters

How to Use

1. Add reference images — Provide clear pictures of the subjects or elements you want to use.
2. Write your prompt — Explain what to change and what to retain.
3. Choose the output settings — Optionally set size, then choose JPEG or PNG.
4. Generate — Submit your request and download the result.

Pricing

$0.035 per image, regardless of the supported output size or output format. Reference images do not add separate charges.

Best Use Cases

• Style exploration — Reimagine existing images in different visual styles.
• Product concepts — Explore settings, materials, and presentation.
• Creative compositing — Use multiple references to guide a new composition.
• Marketing visuals — Adapt images for campaigns and social content.

Pro Tips

• Use clear reference images and describe their roles in the prompt.
• Refer to images by their upload order, such as “Figure 1” and “Figure 2”.
• Specify both the changes you want and the details you want to retain.

Notes

• Both prompt and images are required. Supply image URLs that the service can access.
• Omit size for automatic sizing; this does not guarantee the same pixel dimensions as the reference image.
• Custom sizes must contain between 3,686,400 and 16,777,216 total pixels, with a width-to-height ratio between 1:16 and 16:1. Both conditions must be met.
• Total pixels are calculated as width × height. Not every combination shown by the size controls meets these limits.
• Specify dimensions through size; separate width and height parameters are not supported.

Related Models

• Seedream V5.0 Lite Text-to-Image — Generate a single image from a text prompt.
• Seedream V5.0 Lite Sequential — Generate multiple images from one prompt.
• Seedream V5.0 Lite Edit Sequential — Create multiple images guided by reference pictures.

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'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "images": [
    "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg"
  ],
  "output_format": "jpeg"
}
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/bytedance/seedream-v5.0-lite/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

Parameters

Task Submission Parameters

Request Parameters

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

Response Parameters

Result Request Parameters

Result Response Parameters