Muapi에서 Use Seedance 2.5 Motion Control in international mode to recast performers, preserve choreography, camera work and location, and render 4–30s video. Try free.
Muapi에서 Use Seedance 2.5 Motion Control in international mode to recast performers, preserve choreography, camera work and location, and render 4–30s video. Try free.
이 모델 정보
Muapi의 Seedance 2.5 Intl Motion Control recasts the performers in an existing performance, dance, or UGC video while preserving the original motion, choreography, blocking, camera work, timing, and location. Provide a source video plus up to 30 character reference images and an optional instruction. Muapi automatically extracts motion, cleans representative environment frames, prepares character references, and renders the final 720p recast through the international-region Seedance 2.5 path. The request shape matches the standard motion-control endpoint, while this variant uses the international deployment and carries a 10% tier surcharge.
비용 분석
| 공급자 | 비용 | 메모 |
|---|---|---|
| muapiapp | Standard 2.5 Motion Control price × 1.1 | 요금 안내: One asynchronous request covers preprocessing and the final 720p recast; source-video depth extraction remains duration-based. |
| Fal.ai | 제공되지 않음 | 요금 안내: No equivalent single-request international motion-recast pipeline is listed. |
| Replicate | 제공되지 않음 | 요금 안내: No equivalent single-request international motion-recast pipeline is listed. |
요금 안내: One asynchronous request covers preprocessing and the final 720p recast; source-video depth extraction remains duration-based.
요금 안내: No equivalent single-request international motion-recast pipeline is listed.
요금 안내: No equivalent single-request international motion-recast pipeline is listed.
** 경쟁업체 가격은 유사한 모델 아키텍처 및 사용 계층을 기준으로 추정됩니다.
구성 스키마
| 매개변수 | 유형 | 설명 | 기본값 |
|---|---|---|---|
| 프롬프트 | string | Optional short creative brief describing the recast. If omitted, roles are inferred automatically from the reference video and images. | Swap the two dancers with the people in the reference images, keep the same room. |
| target | string | Reference video whose motion, choreography, blocking, and camera work will be preserved. | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-omni-reference.mp4 |
| Character targets | array | Up to 30 character, location, or style reference images used to recast the performers and environment. | https://cdn.muapi.ai/assets/image_download_21.avif |
| 종횡비 | 열거형(8 옵션) | 출력 비디오 종횡비. | 16:9 |
| 기간(초) | int | Video duration in seconds (4-30). | 5 |
| 종자 | int | 씨앗. -1. | 42 |
| 높은 비트 전송률 | boolean | , . | false |
| 오디오 생성 | boolean | Generate audio (music/SFX) for the recast video. Off by default. | false |
Optional short creative brief describing the recast. If omitted, roles are inferred automatically from the reference video and images.
Swap the two dancers with the people in the reference images, keep the same room.Reference video whose motion, choreography, blocking, and camera work will be preserved.
https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/seedance-v2.0-omni-reference.mp4Up to 30 character, location, or style reference images used to recast the performers and environment.
https://cdn.muapi.ai/assets/image_download_21.avif출력 비디오 종횡비.
16:9Video duration in seconds (4-30).
5씨앗. -1.
42, .
falseGenerate audio (music/SFX) for the recast video. Off by default.
false개발자 문서
video_url.images_list; one image per performer is recommended.prompt. Set duration from 4 to 30 seconds, choose aspect_ratio, and optionally set seed or high_bitrate.curl -X POST https://api.muapi.ai/api/v1/seedance-2.5-intl-motion-control \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"video_url": "https://example.com/reference-performance.mp4",
"images_list": ["https://example.com/performer.jpg"],
"prompt": "Recast the performer and preserve the choreography and location.",
"duration": 5,
"aspect_ratio": "16:9"
}'
GET /api/v1/predictions/{request_id}/result or provide a webhook URL to receive the completed video.자주 묻는 질문
It preserves the source video's motion, choreography, blocking, timing, camera work, and filming location while recasting the performers from your reference images.
The request and staged preprocessing pipeline are the same, but Intl selects the international-region Seedance 2.5 deployment and is priced 10% above the standard 2.5 motion-control tier.
You can provide up to 30 URLs in `images_list`. For multiple performers, provide one clear character reference image per performer in the desired mapping order.
No. `prompt` is optional; when omitted, the pipeline infers performer roles and the recast from the reference video and supplied images.
Duration can be 4–30 seconds. Supported aspect ratios are adaptive, 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, and 9:21; output is rendered at 720p.
The bill combines the standard 2.5 motion-control render and preprocessing charges with a 10% Intl surcharge. Preprocessing includes depth extraction, environment cleanup, prompt analysis, and character-sheet preparation.