Bytedance Seedance V1 Lite T2V 720p API Documentation

Bytedance Seedance V1 Lite T2V 720p API Documentation

Playground

Seedance V1 Lite produces coherent multi-shot 720p videos with smooth motion and accurate following of detailed text prompts. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

> Notice — Service Stability > > The Seedance 1.0 family is being retired or has already been retired by its original provider. Until it is fully removed, the service is unstable: generations may fall back to alternative models without notice, results can differ from what this model used to produce, and the service can be temporarily unavailable. > > If you need an equally capable, stable alternative, we recommend Seedance 2: bytedance/seedance-2.0/text-to-video.

Seedance v1 Lite T2V 720p generates short videos directly from a text prompt. Describe the subject, action, scene, and camera intent, and the model produces a coherent 720p clip with smooth motion—great for story beats, anime-style moments, quick pre-visualization, and rapid creative iteration. You can optionally enable camera_fixed to keep framing steady while motion happens within the scene.

Key capabilities

• Text-to-video generation (720p) optimized for fast iteration
• Good motion from clear action verbs and scene direction
• Optional camera_fixed mode for stable framing (no camera movement)
• Supports multiple aspect ratios for common layouts (e.g., 16:9)

Use cases

• Short story moments and animated shots (5–20s beats)
• Anime-style scene generation with expressive lighting and motion
• Quick pre-vis for camera blocking and action timing
• Social-ready clips and concept trailers at lightweight cost
• Creating variations by changing seed while keeping the same prompt

Pricing

Parameters

• prompt (required): Describe what happens in the video (subject, action, scene, mood)
• aspect_ratio: Output aspect ratio (e.g., 16:9, 9:16, 1:1)
• duration (required): Video length in seconds
• camera_fixed: Whether to fix the camera position
• seed: Random seed (-1 for random; fixed value for reproducible results)

Prompting guide (T2V)

Write prompts like a director’s brief:

• Subject: who/what is on screen
• Action: what changes over time (fluttering, turning, walking, reacting)
• Scene: where it happens + time of day
• Lighting: sunset glow, neon, soft volumetrics, etc.
• Camera: only describe movement if camera_fixed is off

Example pattern: A [subject] [action] in [scene], [lighting], [mood], smooth motion, cinematic pacing.

Example prompts

• A high school girl stands at a rooftop fence during sunset, hair and skirt fluttering in the wind, orange sky glowing behind her, distant courtyard ambience, soft anime-style shading, vivid colors, gentle camera drift.
• Close-up of a musician’s hands playing guitar, warm indoor lighting, shallow depth of field, subtle handheld motion, realistic details.
• A sci-fi corridor scene with drifting fog and blinking lights, a character slowly turns toward the camera, dramatic rim light, cinematic tension.

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",
  "duration": 5,
  "aspect_ratio": "16:9",
  "camera_fixed": false
}
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/seedance-v1-lite-t2v-720p" \
  -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