Seedance 2 首尾帧 Fast: Image-to-Video
ByteDance 的 SD 2 首尾帧（Fast）。以更低成本快速生成在参考图像之间过渡的视频。提供 1 张或 2 张图像。
ByteDance 的 SD 2 首尾帧（Fast）。以更低成本快速生成在参考图像之间过渡的视频。提供 1 张或 2 张图像。
About this model
SD 2 首尾帧（Fast）以更低成本快速生成在参考图像之间过渡的视频。提供 1 张或 2 张图像来锚定开场帧和结束帧，支持 4–15 秒时长。
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.15 per second (e.g. $0.75 for 5 seconds) | 생성된 영상 초당 요금 청구. |
| Fal.ai | $0.3024/sec (high) / $0.2419/sec (basic) | Fal.ai charges $0.3024/sec for high quality and $0.2419/sec for basic. muapiapp Fast at $0.15/sec is 50% cheaper than Fal.ai's basic rate. |
| Replicate | $0.3024/sec (high) / $0.2419/sec (basic) | Replicate charges the same as Fal.ai — $0.3024/sec (high), $0.2419/sec (basic). muapiapp Fast at $0.15/sec is 50% cheaper than Replicate's basic rate. |
생성된 영상 초당 요금 청구.
Fal.ai charges $0.3024/sec for high quality and $0.2419/sec for basic. muapiapp Fast at $0.15/sec is 50% cheaper than Fal.ai's basic rate.
Replicate charges the same as Fal.ai — $0.3024/sec (high), $0.2419/sec (basic). muapiapp Fast at $0.15/sec is 50% cheaper than Replicate's basic rate.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| 프롬프트 | string | 指导帧间过渡的文本描述。 | Two people having a street interview, the interviewer holds a microphone. |
| 帧이미지 | array | 1 张이미지仅作为首帧；2 张이미지作为首帧和末帧。使用 'adaptive' 화면 비율以匹配参考이미지几何结构。 | https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg |
| 화면 비율 | Enum (7 options) | 출력비디오화면 비율。'adaptive' 会匹配参考이미지（推荐）；具体比例可能会裁剪或补边。 | adaptive |
| 재생 시간 (초)（秒） | int | 비디오재생 시간（秒）。 | 5 |
指导帧间过渡的文本描述。
Two people having a street interview, the interviewer holds a microphone.1 张이미지仅作为首帧；2 张이미지作为首帧和末帧。使用 'adaptive' 화면 비율以匹配参考이미지几何结构。
https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/712345784292/4a8c5c70-abcc-4920-873e-b0e219986453.jpg출력비디오화면 비율。'adaptive' 会匹配参考이미지（推荐）；具体比例可能会裁剪或补边。
adaptive비디오재생 시간（秒）。
5Developer documentation
上传 1 或 2 张图像：通过 images_list 提供图像。第一张图像 = 开场帧，第二张图像（可选）= 结束帧。
编写프롬프트：描述帧之间的转场或运动。
设置时长：设置为 4 到 15 秒。
提交并轮询：使用返回的 request_id 轮询结果。
Frequently asked
不需要。1 张图像仅锚定第一帧；2 张图像会同时约束视频的起始帧和结束帧。
Fast 版本速度更快、成本更低，但与 Pro 版本相比，출력 결과质量会略有下降。
支持 JPEG、PNG、WebP 和 BMP。