Bytedance Seedance V1 Pro I2V 1080p API Documentation

Bytedance Seedance V1 Pro I2V 1080p API Documentation

Playground

Seedance V1 Pro generates coherent multi-shot image-to-video outputs with smooth motion and prompt-accurate results in 1080p. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Seedance v1 Pro I2V 1080p generates high-quality 1080p videos from a single reference image plus a motion-focused prompt. Upload a starting image, describe the action and camera intent, and the model animates the scene while keeping the input image as the visual anchor. You can also provide an optional last_image to better control how the clip ends, and enable camera_fixed to keep the camera locked while motion happens within the frame.

Key capabilities

• Image-to-video generation anchored to a reference image (1080p output)
• Natural motion and smooth animation for character actions and scene beats
• Optional last_image support for improved ending continuity
• Optional camera_fixed mode for fixed framing (no camera movement)
• Seed control for reproducible iterations

Use cases

• High-resolution character animation from a single keyframe
• Cinematic scene beats with controlled blocking and fixed shots
• Start-to-end guided shots using last_image for more predictable endings
• Marketing and storytelling clips that need sharper detail at 1080p
• Generating multiple takes by changing seed while keeping the same prompt

Pricing

Inputs

• image (required): starting reference image (first-frame anchor)
• prompt (required): motion and scene direction
• last_image (optional): end-frame reference to guide the finishing state

Parameters

• prompt: describes subject action, scene motion, and camera direction
• image: input image (upload or URL)
• last_image: optional end reference image (upload or URL)
• duration: video length in seconds
• aspect_ratio: output aspect ratio selection
• camera_fixed: whether to fix the camera position
• seed: random seed (-1 for random; fixed value for reproducible results)
• safety checker: enabled filtering for safer generations (if available in your UI)

Prompting guide (I2V)

For better control, write prompts like a director’s beat sheet:

• Blocking: who moves where (walks from right to left, stops, gestures)
• Emotion: facial expression and body language cues (angry, disappointed, calm)
• Camera: fixed shot, slow push-in, pan, orbit (or keep camera_fixed on)
• Continuity: keep identity, outfits, and scene layout consistent with the input image

Example prompts

• Fixed shot. The man on the right walks toward the man on the left and confronts him, visibly dissatisfied. Strong body language, tense atmosphere, natural motion, realistic lighting, 5 seconds.
• A character turns, takes two steps forward, and points while speaking, cinematic lighting, subtle wind and particles, camera fixed, 5 seconds.
• Slow push-in on a hero’s face as their expression changes from calm to determined, shallow depth of field, cinematic mood, 5 seconds.

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",
  "duration": 5,
  "aspect_ratio": "21:9",
  "camera_fixed": false,
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/bytedance/seedance-v1-pro-i2v-1080p" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters