Bytedance Seedream V4 Edit Sequential API Documentation

Bytedance Seedream V4 Edit Sequential API Documentation

Playground

Seedream 4.0: 4K image generation and editing with character and object consistency and sequential multi-image outputs. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

What is it?

Seedream v4 Edit Sequential is image-to-image model for editing a source image and generating a sequence/group of consistent outputs in one pass. It’s built for high feature retention (identity, logos, layout) and cross-image style continuity—ideal for matched sets, story panels, or product variants.

What makes it stand out?

• I2I + sequential in one pipeline – Edit a single source and produce N coherent images (series/panels) without switching models.
• Five core strengths – Precise instruction editing, high feature retention, deep scene understanding, ultra-fast inference, ultra-high-res output.
• Consistency controls – Strong identity/style preservation across all images in the set.
• Rich edit ops – Add/remove elements, attribute/style change, structural tweaks (e.g., pose/face swap), texture/brush/frame edits.

Designed for

• Commercial design — Posters, apparel, packaging, e-commerce sets; fast, brand-safe varianting.
• Entertainment IP — Character look-dev, key art sequences with locked identity.
• Fine art & Illustration — Multi-piece series with coherent palette and linework.
• Architecture — Material/lighting variations across consistent viewpoints.
• Brand content — Campaign sets and social carousels with fixed logo/palette.

How to use

1. Inputs: Upload several source images.
2. Prompt: Write the instruction and repeat the same N in text (e.g., “a series of N images / Panels 1–N”) to lock count & continuity.
3. Count: Set max_images = N for the number of images.
4. Control the size: the max size is 8192 * 8192.
5. Generate → review → iterate (reuse or change seed for A/B).

Please Note: Declare the number of images twice — max_images = N and inside the prompt!

Price

• $0.027 per image.
• Total price = max_images * $0.027

Prompting guide

• Edit instruction (per set) Replace [object A] with [object B]; keep [logo/identity/features]; preserve [lighting/style].
• Sequential consistency (count locked) Generate a series of [N] edited images (Panels 1–[N]) from the source, maintaining the same [character/product/logo] identity, palette, and composition style.

Panel 1 — [edit/shot] Panel 2 — [edit/shot] … Panel N — [edit/shot]

• Terminology Use precise, domain-native terms (photography, fashion, architecture) to match expectations.

Example (product variant set, N=4)

Set max_images = 4 and use:

Generate a series of 4 edited images (Panels 1–4) from the source photo, keeping the same shoe model and logo placement. Maintain identical angle, lighting, and background; change only the colorway per panel:

Panel 1 — classic white + black swoosh

Panel 2 — navy + gold accents

Panel 3 — matte red + white outsole

Panel 4 — forest green + gum sole

Ensure consistent proportions, stitching detail, and material texture across all 4 panels.

Note

Please set the max_image first, and then input how many images you want to generate in prompt! Such as:

• max_image = 4.
• Prompt: I want to generate 4 images… + (your prompt)

Recommended Resolutions

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"
  ],
  "max_images": 1
}
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-v4/edit-sequential" \
  -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