Video Merger API Documentation

Video Merger API Documentation

Playground

Try it on WaveSpeedAI!

Video Merger joins 2 to 20 video clips into one video with crossfades or straight cuts and optional background music, fitting every clip into the first clip’s frame. Ready-to-use REST inference API, no coldstarts, affordable pricing.

Features

Video Merger joins 2 to 20 video clips into one video, in the order you give them, up to 600 seconds in total. Use crossfades or straight cuts, add background music, and get a single MP4 ready to post. It is the last step of any multi-shot video: the shots of a short film, the episodes of a UGC series, or several hooks in front of one ad.


How to Use

  1. Add your clips in playing order (2 to 20 videos). The merged video is cut at 600 seconds.
  2. Choose a transition: fade crossfades picture and sound between clips, none cuts straight from one clip to the next. Set transition_duration for the length of each crossfade.
  3. Add music (optional): give audio and its audio_volume. The music is looped or cut to the length of the video and fades out at the end.
  4. Keep or drop the clips’ sound with keep_original_audio.
  5. Submit and download the merged video.

Pricing

Billing is per second of the clips’ total length (each started second counts), with a minimum of 10 seconds and a maximum of 600 seconds.

ClipsPrice
2 clips of 5 seconds (10 seconds)$0.01
3 clips of 10 seconds (30 seconds)$0.03
4 clips of 15 seconds (60 seconds)$0.06
10 clips of 30 seconds (300 seconds)$0.30
20 clips of 30 seconds (600 seconds, the maximum billed length)$0.60

Output

  • One MP4 (H.264 video, AAC audio) in the size and frame rate of the first clip.
  • Clips with another shape are fitted into that frame with black bars, never stretched; other frame rates are converted.
  • Clips without sound get silence, so music and crossfades still line up.

Limits

  • 2 to 20 clips. The merged video is at most 600 seconds: clips that would start later are left out and the last one is cut.
  • transition_duration from 0.1 to 2 seconds; a crossfade is never longer than half of the shortest clip.

Tips

  • Generate every shot at the same aspect ratio and resolution for a seamless film.
  • Short crossfades (0.3 to 0.5 seconds) suit story shots; use none for fast-paced ads and hook tests.
  • Keep the music around 0.2 to 0.4 when the clips contain dialogue.

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'
{
  "videos": [
    "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
    "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4"
  ],
  "transition": "fade",
  "transition_duration": 0.5,
  "audio_volume": 0.3,
  "keep_original_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/wavespeed-ai/video-merger" \
  -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

Retry temporary errors only while polling, never the submit request. Errors, retries and timeouts

Parameters

Task Submission Parameters

Request Parameters

ParameterTypeRequiredDefaultRangeDescription
videosarray<string>Yes-2 ~ 20 itemsThe clips to join, in playing order: 2 to 20 videos. The merged video is at most 600 seconds; clips that would start later are left out and the last one is cut. Clips of another shape are fitted into the first clip's frame with black bars, never stretched.
transitionstringNofadefade, nonefade crossfades picture and sound between clips; none cuts straight from one clip to the next.
transition_durationnumberNo0.50.1 ~ 2Length of each crossfade in seconds (used with transition fade).
audiostringNo--Optional background music, looped or cut to the length of the merged video and faded out at the end.
audio_volumenumberNo0.30 ~ 1Volume of the background music, from 0 (silent) to 1 (as recorded).
keep_original_audiobooleanNotrue-Keep the clips' own sound. Turn off to use only the background music, or to get a silent video.

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
data.idstringUnique identifier for the prediction, Task Id
data.modelstringModel ID used for the prediction
data.outputsarrayOutput values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed)
data.urlsobjectObject containing related API endpoints
data.statusstringTask status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses.
data.created_atstringISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds

Result Request Parameters

ParameterTypeRequiredDefaultDescription
idstringYes-Task ID

Result Response Parameters

ParameterTypeDescription
codeintegerHTTP status code (e.g., 200 for success)
messagestringStatus message (e.g., “success”)
dataobjectThe prediction data object containing all details
data.idstringUnique identifier for the prediction
data.modelstringModel ID used for the prediction
data.outputsarray<string | object>Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model.
data.urlsobjectObject containing related API endpoints
data.statusstringStatus: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses
data.created_atstringISO timestamp of when the request was created
data.errorstringError message (empty if no error occurred)
data.timingsobjectObject containing timing details
data.timings.inferenceintegerInference time in milliseconds
© 2026 WaveSpeedAI. All rights reserved.