Bytedance Seedream V4 Edit API Documentation

Bytedance Seedream V4 Edit API Documentation

Playground

Seedream V4 Edit is state-of-the-art image editing model that outperforms Nano Banana in fidelity and edit quality. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Seedream 4.0/Edit is a specialized image-to-image model for accurate edits to existing images - swap outfits and accessories, adjust hair or makeup, recolor or re-materialize products, and replace interior finishes like floors, walls, or furniture while maintaining subject identity, lighting, and overall composition.

Highlights

• High fidelity on people & products: Reliable skin tones, fabric/material details, logos, and fine edges.
• Production-friendly consistency: Generate multiple variants quickly with locked camera/look and brand color grading.
• Structured prompts that scale: Works best when the goal is clear up front (action + object + target feature + constraints).

Use cases

• Portrait & influencer workflows: Outfit/makeup/hair changes, branded KV series, quick campaign variants.
• E-commerce & product teams: Colorways, material/texture swaps, packaging updates.
• Interior/archviz: Wall/floor finish replacements, upholstery changes, lighting-aware set dressing.
• Marketing & growth: Fast A/B testing with consistent brand color and camera setup.

Price

Only $0.027 per image!!!

How to use

1. Prepare source: Upload the base images (Up to 10 images).
2. Set size: The maximum size of the image is 8192 * 8192.
3. Write a clear prompt: Write clearly in the prompt about object, feature, and constraints.

Prompt patterns (copy-ready)

Use: change action + change object + target feature + constraints (keep/avoid)

Portrait (KV series)

portrait KV series, {STYLE} style, consistent color grading {BRAND_COLOR}, fixed camera look (85mm shallow depth), interchangeable persona: {PERSONA}, reserved lower-third text “{NAME} — {ROLE}”

Change Clothes / Jewellery / Makeup

Outfit swap for portrait, replace clothing with {OUTFIT_DESC}; keep pose and composition; accessories {JEWELRY_DESC}; makeup/hair {MAKEUP_HAIR}; preserve skin tone and lighting; clean edges, no artifacts

Background Replacement

Background replacement for subject, keep subject edges; new environment: {SCENE_DESC}; match light direction and color temperature; soft contact shadows; no haloing

Interior / Outdoor Replacement

Interior finish swap, update wall {WALL_MATERIAL}, floor {FLOOR_MATERIAL}, furniture upholstery {FABRIC}; layout and lighting unchanged; realistic PBR textures

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"
  ]
}
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" \
  -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