Bytedance Seedance 2.0 Video Edit Turbo API Documentation

Bytedance Seedance 2.0 Video Edit Turbo API Documentation

Playground

Seedance 2.0 (Video-Edit Turbo) is the turbo tier for editing an input video from a natural-language prompt — faster, more affordable high-resolution output while preserving subject identity, composition, and motion. Ready-to-use REST API, best performance, no coldstarts, affordable pricing.

Features

Seedance 2.0 Video-Edit Turbo is the turbo tier for editing an input video from a natural-language prompt — faster, more affordable high-resolution output while preserving subject identity, composition, and motion of the original.

Key Features

• Turbo tier — Optimized for cost and speed at higher resolutions.
• Conversational video editing — Describe the change in plain language; the model rewrites the scene while keeping the original motion intact.
• Subject and motion preservation — Faces, objects, and camera movement from the input video stay consistent through the edit.
• Multi-reference support — Optionally guide style, character identity, or audio with reference images and audio clips.
• Native audio synchronization — Generates synchronized audio in a single pass.

Parameters

How to Use

1. Upload the input video. Anything longer than 15 s is trimmed to the first 15 s automatically.
2. Write the edit prompt. Describe the change you want — the prefix Edit the input video. is added for you.
3. (Optional) Add references. Reference images can constrain style or identity; reference audio can constrain the soundtrack.
4. Pick a turbo resolution. 720p or 1080p.
5. Run. Receive the edited video at the requested resolution with synchronized audio.

Writing Effective Prompts

Editing keeps your source video and changes only what you name. Be precise about the scope, and describe the change as from A to B.

State the change plainly

Name exactly what to add, remove, or modify — and use an editing trigger word (add, remove, replace, change to, modify) so the intent is unambiguous.

• “Replace the man’s coffee cup with a book, keep everything else unchanged.”
• “Remove the subtitles.”
• “Change the background from a city street to a snowy forest.”

Use timestamps for partial edits

Confine a change to a time range so the rest of the clip is untouched.

• “From 4-6s, change the man’s action from drinking coffee to waving; leave the rest unchanged.”

Reference images for edits

Attach images to specify a replacement, and bind each explicitly (see below).

• “Replace the woman on the right with the person in @Image 1.”

Referencing uploaded assets

When you attach images, videos, or audio, bind each one explicitly in the prompt by its upload order — @Image 1, @Video 1, @Audio 1 — and say what it’s for. Don’t rely on labels drawn inside the image itself.

• “The knight in @Image 1 walks through the castle in @Image 2.”
• “Refer to @Video 1 for the camera movement only; keep its shot order.”
• “@Image 1 and @Image 2 are Character 1, voiced by @Audio 1.”

When a reference is already accurate, just point to it — no need to re-describe it in detail.

Audio edits

You can add, remove, or modify audio too.

• “Translate the dialogue to Spanish, keep lip movements matched, no subtitles.”
• “Remove the background music; keep only ambient and action sounds.”

Negative control

Positive descriptions work best, but you can suppress subtitles and audio:

• “No subtitles.”
• “No BGM — ambient and action sounds only.”
• “No audio.”

Weak vs. strong

Pricing

Billed as a base rate on input duration + output duration seconds, plus a resolution surcharge on output seconds only. Input duration is clamped to the 2-15 s range (sources shorter than 2 s are padded with the last frame).

Examples (input 5 s, output 5 s):

Examples (input 12 s, output 12 s):

Notes

• Inputs longer than 15 s are trimmed to 15 s and inputs shorter than 2 s are padded with the last frame to 2 s before editing — billing reflects the conformed length.
• Auto-detected duration matches the input length (rounded up), clamped to the 4-15 s range.
• Native audio generation is included.

Related Models

• Seedance 2.0 Video-Edit — Standard tier at 480p / 720p / 1080p.
• Seedance 2.0 Fast Video-Edit Turbo — Faster, cheaper turbo tier.

Notice: Use @Image 1, @Image 2, @Audio 1, etc. to reference your uploaded assets. Number each type separately, in upload order.

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",
  "video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
  "aspect_ratio": "16:9",
  "resolution": "720p",
  "enable_web_search": false,
  "generate_audio": true
}
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-2.0/video-edit-turbo" \
  -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

Response Parameters

Result Request Parameters

Result Response Parameters