创建 HappyHorse 视频

POST

videos

curl https://api.cometapi.com/v1/videos \
  -H "Authorization: Bearer $COMETAPI_KEY" \
  -F model=happyhorse-1.1 \
  -F 'prompt=A close-up cinematic shot of a ceramic coffee cup on a wooden table. Steam rises slowly while morning light moves across the background.' \
  -F seconds=3 \
  -F size=1920x1080

{
  "id": "task_example",
  "task_id": "task_example",
  "object": "video",
  "model": "happyhorse-1.1",
  "status": "queued",
  "progress": 0,
  "created_at": 1779938152
}

根据文本、参考图像或首帧创建 HappyHorse 视频。API 会立即返回任务 ID。请存储返回的 id，以便轮询任务并下载完成的视频。 POST /v1/videos 使用 multipart/form-data。将控制参数作为表单字段发送，并将参考图像作为 input_reference 字段发送，每个字段对应一个 HTTPS URL 或上传的图像文件。

选择输入模式

目标	模型	必填字段	可选字段
文生视频	`happyhorse-1.0`, `happyhorse-1.1`	`model`, `prompt`	`seconds`, `size`
参考图生视频	`happyhorse-1.0`, `happyhorse-1.1`	`model`, `prompt`, 1–9 `input_reference` URL 或文件	`seconds`, `size`
首帧图生视频	`happyhorse-1.0`, `happyhorse-1.1`	`model`, `prompt`，一个 `first_frame` URL 或文件	`seconds`, `size`

参考图像可引导视频中的主体和视觉细节。发送 1–9 个可公开访问的 HTTPS 图像 URL 或 1–9 个图像文件，按图像顺序为每张图像重复发送一次 input_reference。在此字段中只能使用 URL 或只能使用文件。请分别发送每个 URL，而不是用逗号连接多个 URL。对于 input_reference，每张参考图像必须不超过 10 MB，且需满足 happyhorse-1.0。使用 happyhorse-1.1 时，每张参考图像必须不超过 20 MB。这些模型限制同时适用于图像 URL 和上传的文件。请使用 JPEG、JPG、PNG 或 WEBP 图像，且最短边至少为 400 px。每个文件的 input_reference 上传限制为 20 MiB（20 × 1024 × 1024 字节）。上传的图像还必须满足上述模型专属限制。本页顶部的请求示例展示了 HappyHorse 1.1 的多个图像 URL，以及两个模型的多个文件上传。要设置首帧，请使用一个可公开访问的 HTTPS 图像 URL 或一个图像文件发送一次 first_frame。请在与 input_reference 分开的请求中使用此字段。

设置时长和尺寸

模型 ID	`seconds`	默认值
`happyhorse-1.0`	整数 `3`-`15`	`5` 秒， `1920x1080`
`happyhorse-1.1`	整数 `3`-`15`	`5` 秒， `1920x1080`

请从下表中为 WxH 选择一个精确的 size 值。对于首帧请求，图像决定画面构图和宽高比，而 size 用于选择输出分辨率级别。

模型	画面形状	`size` 取值
`happyhorse-1.0`	横向	`1280x720`, `1920x1080`
	纵向	`720x1280`, `1080x1920`
	方形	`1440x1440`
`happyhorse-1.1`	横向	`1280x720`, `1920x1080`
	纵向	`720x1280`, `1080x1920`
	方形	`1440x1440`
	经典横向	`1440x1080`
	经典纵向	`720x960`, `1080x1440`
	宽幅横向	`2520x1080`
	宽幅纵向	`1080x2520`
	近方形横向	`900x720`, `1350x1080`
	近方形纵向	`720x900`, `1080x1350`

任务流程

创建任务

发送 multipart 表单请求，并存储返回的 id。

轮询任务

调用获取 HappyHorse 视频直到 status 为 completed、failed 或 error。

下载结果

当任务处于 completed 时，调用获取 HappyHorse 视频内容以下载 MP4 文件。

授权

Authorization

string

header

必填

Bearer authentication. Use your CometAPI API key.

请求体

multipart/form-data

model

enum<string>

必填

HappyHorse model ID for this endpoint.

可用选项:

happyhorse-1.0,

happyhorse-1.1

示例:

"happyhorse-1.1"

prompt

string

必填

Text prompt that describes the video to generate. For image inputs, describe the motion and visual details to preserve.

示例:

"Use the uploaded reference image as the visual guide. Create a short cinematic shot with gentle natural motion."

seconds

integer

默认值:5

Requested clip duration in whole seconds. Send an integer from 3 through 15 as a form field. The default is 5.

必填范围: 3 <= x <= 15

示例:

3

size

string

Output size as an exact WxH value. Use one of the documented size values for the selected HappyHorse model. For first-frame requests, size selects the resolution tier; the image determines the framing and aspect ratio.

示例:

"1920x1080"

input_reference

Reference images for reference-to-video. Send 1–9 publicly accessible HTTPS image URLs or 1–9 uploaded image files by repeating input_reference once per image. Use only URLs or only files within this field; do not join URLs with commas. Reference images guide subjects and visual details. CometAPI preserves image order. Use JPEG, JPG, PNG, or WEBP images with a shortest side of at least 400 px. Each reference image must be at or below 10 MB for happyhorse-1.0 or 20 MB for happyhorse-1.1. These model limits apply to both image URLs and uploaded files. The input_reference file upload limit is 20 MiB (20 x 1024 x 1024 bytes) per file. Keep uploads within the model limit as well. Keep this field separate from first_frame.

Pattern: ^https://

first_frame

Opening-frame image as one publicly accessible HTTPS URL or one image file. Send this field once, using either a URL or a file. Use JPEG, JPG, PNG, or WEBP images up to 20 MB. Do not combine first_frame with input_reference.

Pattern: ^https://

响应

Task created. Store the returned id and poll GET /v1/videos/{task_id}.

string

必填

Task ID. Use this value with retrieve and content endpoints.

示例:

"task_example"

object

string

必填

Object type. Video tasks return video.

示例:

"video"

model

string

必填

Model ID used for the task.

示例:

"happyhorse-1.1"

status

enum<string>

必填

Task lifecycle status. Poll until the value is completed, failed, or error.

可用选项:

queued,

in_progress,

completed,

failed,

error

示例:

"queued"

progress

integer

必填

Task progress as a coarse percentage.

必填范围: 0 <= x <= 100

示例:

0

created_at

integer

必填

Task creation time as a Unix timestamp in seconds.

示例:

1779938152

task_id

string

Compatibility alias for id when present.

示例:

"task_example"

completed_at

integer

Task completion time as a Unix timestamp in seconds. This field appears on completed tasks.

示例:

1779938219

video_url

string

Temporary video delivery URL. This field appears on completed tasks.

示例:

"<temporary-video-url>"

error

object

Failure details. This field appears when the task fails.

最后修改于 2026年9月9日

curl https://api.cometapi.com/v1/videos \
  -H "Authorization: Bearer $COMETAPI_KEY" \
  -F model=happyhorse-1.1 \
  -F 'prompt=A close-up cinematic shot of a ceramic coffee cup on a wooden table. Steam rises slowly while morning light moves across the background.' \
  -F seconds=3 \
  -F size=1920x1080

{
  "id": "task_example",
  "task_id": "task_example",
  "object": "video",
  "model": "happyhorse-1.1",
  "status": "queued",
  "progress": 0,
  "created_at": 1779938152
}

内容审核

API 密钥

选择输入模式

设置时长和尺寸

任务流程

授权

请求体

响应

​选择输入模式

​设置时长和尺寸

​任务流程

授权

请求体

响应

选择输入模式

设置时长和尺寸

任务流程