创建 Flux 3 视频

POST

videos

curl \
  https://api.cometapi.com/v1/videos \
  -H "Authorization: Bearer $COMETAPI_KEY" \
  --form-string 'model=flux-3' \
  --form-string 'prompt=A paper boat glides across a still pond while the camera moves forward.' \
  --form-string 'seconds=5' \
  --form-string 'size=1280x720'

{
  "id": "<task_id>",
  "task_id": "<task_id>",
  "object": "video",
  "model": "flux-3",
  "status": "queued",
  "progress": 0,
  "created_at": 1779938152
}

通过文本、参考图像或首尾帧创建 Flux 3 视频。保存返回的 id，以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。对于图像输入，请使用参考图像 URL 或起始帧和结束帧。将 input_reference 与 first_frame 和 last_frame 分开。

选择输入模式

目标	必填字段	可选字段
文生视频	`model=flux-3`, `prompt`	`seconds`, `size`
HTTPS 图生视频	`model=flux-3`, `prompt`，一个 `input_reference` URL	`seconds`, `size`
有序关键帧	`model=flux-3`, `prompt`，重复的 `input_reference` URL	`seconds`, `size`
上传起始帧	`model=flux-3`, `prompt`，一个 `first_frame` 文件	`seconds`, `size`
首尾帧视频	`model=flux-3`, `prompt`, `first_frame`, `last_frame`	`seconds`, `size`

使用参考图像

input_reference 字段仅接受可公开访问的 HTTPS 图像 URL。对于单张图像，发送一次该字段；对于多张图像，则按照关键帧顺序重复发送。每个字段使用一个 URL，最多共 10 个 URL。描述生成的视频应从参考图像中保留的运动、相机行为和视觉细节。本页顶部的请求示例展示了单图和多图输入。以下请求按提交顺序使用两张图像：

curl \
  https://api.cometapi.com/v1/videos \
  -H "Authorization: Bearer $COMETAPI_KEY" \
  --form-string 'model=flux-3' \
  --form-string 'prompt=Move smoothly through the supplied reference images in their given order. Preserve recognizable subjects and visual details as the scene changes. Keep the camera movement gentle.' \
  --form-string 'seconds=5' \
  --form-string 'size=1280x720' \
  --form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569429.png' \
  --form-string 'input_reference=https://apidoc.cometapi.com/images/image/gemini/6100640_569420.png'

设置首尾帧

发送 first_frame 以设置起始图像。要同时设置结束图像，请包含 last_frame，并在 prompt 中描述两帧之间的运动。如果仅设置起始帧，请省略 last_frame。结束帧需要同时提供 first_frame 和 last_frame。每个字段接受一个可公开访问的 HTTPS URL 或一个 PNG 或 JPEG 文件。每个字段仅发送一次，使用 URL 或文件之一。每个文件大小不得超过 20 MiB (20 × 1024 × 1024 字节). 本页顶部的帧示例展示了 URL 输入和文件上传，并显式设置了 seconds 和 size。

设置时长和尺寸

将 seconds 显式设置为从 5 到 20 的整数。将 size 设置为以下精确的 WxH 预设值之一：

分辨率档位	尺寸
`720p`	`1280x720`
`1080p`	`1920x1080`

尺寸值用于选择输出分辨率档位。对于文生视频，编码尺寸可为编解码器对齐而调整。对于图生视频，参考图像也可决定最终构图和宽高比。

任务流程

创建任务

发送 multipart 表单请求，并保存返回的 id。

轮询任务

调用获取 Flux 3 视频直到 status 为 completed 或 failed。

下载结果

当任务处于 completed 状态时，调用下载 Flux 3 视频内容以保存 MP4 文件。

授权

Authorization

string

header

必填

Bearer authentication. Use your CometAPI API key.

请求体

multipart/form-data

Flux 3 multipart request. Choose text, input_reference URLs, or first_frame and last_frame inputs. Keep reference-image and frame inputs separate.

model

string

默认值:flux-3

必填

Model ID for this route. Use flux-3.

Allowed value: "flux-3"

prompt

string

默认值:A paper boat glides across a still pond while the camera moves forward.

必填

Text that describes the scene, motion, camera behavior, and visual details that the video should preserve.

Minimum string length: 1

seconds

integer

Requested clip duration in whole seconds. Set seconds explicitly to an integer from 5 through 20.

必填范围: 5 <= x <= 20

size

enum<string>

默认值:1280x720

Resolution preset in exact WxH form. Use 1280x720 for the 720p tier or 1920x1080 for the 1080p tier. Encoded dimensions can be codec-aligned; image-to-video framing can follow the reference image.

可用选项:

1280x720,

1920x1080

input_reference

Reference image URLs. Send one publicly accessible HTTPS image URL per input_reference field. For multiple images, repeat the field in keyframe order, up to 10 URLs. This field accepts URLs only. Keep it separate from first_frame and last_frame.

Pattern: ^https://

first_frame

Opening-frame image as one publicly accessible HTTPS URL or one PNG or JPEG file. Keep each file at or below 20 MiB (20 × 1024 × 1024 bytes). Send the field once, using either a URL or a file. Pair with last_frame to set both boundaries. Keep frame inputs separate from input_reference.

Pattern: ^https://

last_frame

Closing-frame image as one publicly accessible HTTPS URL or one PNG or JPEG file. Keep each file at or below 20 MiB (20 × 1024 × 1024 bytes). Send the field once, using either a URL or a file. To specify the closing image through this endpoint, send last_frame together with first_frame. Keep frame inputs separate from 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 as task_id in retrieve and content requests.

示例:

"<task_id>"

object

string

必填

Object type for the asynchronous video task.

Allowed value: "video"

model

string

必填

Model ID that the task uses.

Allowed value: "flux-3"

status

enum<string>

必填

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

可用选项:

queued,

in_progress,

completed,

failed

progress

integer

必填

Task progress as a coarse percentage.

必填范围: 0 <= x <= 100

created_at

integer<int64>

必填

Task creation time as a Unix timestamp in seconds.

task_id

string

Compatibility alias for id. This field can be omitted from retrieve responses.

示例:

"<task_id>"

completed_at

integer<int64>

Task completion time as a Unix timestamp in seconds when the task provides one.

expires_at

integer<int64>

Result expiration time as a Unix timestamp in seconds when the task provides one.

video_url

string<uri>

Video delivery URL. This field appears on completed tasks.

示例:

"https://media.example.com/flux-3-result.mp4"

error

object

Failure details. This field appears when the task fails.

Show child attributes

最后修改于 2026年9月9日

curl \
  https://api.cometapi.com/v1/videos \
  -H "Authorization: Bearer $COMETAPI_KEY" \
  --form-string 'model=flux-3' \
  --form-string 'prompt=A paper boat glides across a still pond while the camera moves forward.' \
  --form-string 'seconds=5' \
  --form-string 'size=1280x720'

{
  "id": "<task_id>",
  "task_id": "<task_id>",
  "object": "video",
  "model": "flux-3",
  "status": "queued",
  "progress": 0,
  "created_at": 1779938152
}

内容审核

API 密钥

选择输入模式

使用参考图像

设置首尾帧

设置时长和尺寸

任务流程

授权

请求体

响应

​选择输入模式

​使用参考图像

​设置首尾帧

​设置时长和尺寸

​任务流程

授权

请求体

响应

选择输入模式

使用参考图像

设置首尾帧

设置时长和尺寸

任务流程