Videos API: Create Video Generation Task (Standard and Fast)

POST /v1/videos/videos — Generate 4–15s 720p videos with the Videos or Videos Fast model. Supports text, first/last frame, and multi-modal reference.

Submit an asynchronous video generation task using the Videos model series. The endpoint returns a task id immediately — you then poll the Query Video Task endpoint until the task reaches completed or failed. The Videos series supports pure text-to-video, first/last frame guidance, and a powerful multi-modal reference mode combining image and video inputs.

Base URL

https://zcbservice.aizfw.cn/kyyReactApiServer

Endpoint

POST /v1/videos/videos

Authentication

Authorization: Bearer YOUR_API_KEY

Models

Model	Billing	Duration	Resolution	Best For
videos	Per call	4–15s	720p	Standard quality output
videos_fast	Per call	4–15s	720p	Speed-sensitive workflows

Use videos for the highest-quality output. Use videos_fast when turnaround speed matters more than maximum quality.

Request Parameters

model · string · required

The model to use for generation.

Supported values:

• videos — standard model
• videos_fast — faster generation

prompt · string · required

Text description of the video content. Maximum 1500 characters.

Include subject, action, camera movement, and visual style. Avoid content that violates platform policies (e.g., restricted, infringing, or sensitive subject matter).

Example: "A cute kitten playing on a sunlit meadow, cinematic lens"

duration · integer · required

Output video duration in seconds. Required.

• Supported range: 4–15 seconds

ratio · string

Output aspect ratio. Defaults to 16:9.

Value	Description
16:9	Landscape (default)
9:16	Portrait
1:1	Square

resolution · string

Output resolution. Currently only 720p is supported.

Default: 720p

First / Last Frame Mode

Use these fields to guide the video’s starting and ending frames. Both fields are required together.

first_image · string

URL of the first-frame image.

• Required to be used with last_image
• Cannot be combined with referenceImages or referenceVideos

last_image · string

URL of the last-frame image.

• Required to be used with first_image
• Cannot be combined with referenceImages or referenceVideos

First/last frame mode and reference media mode are mutually exclusive. You cannot use first_image/last_image together with referenceImages or referenceVideos in the same request.

Reference Media Mode

Use these fields to provide reference images and/or videos that guide the generated output. They can be used individually or combined.

referenceImages · array

Array of reference image URLs.

• Maximum 4 images
• Each image must be under 20 MB
• Can be combined with referenceVideos for mixed-media reference
• Cannot be used with first_image or last_image

referenceVideos · array

Array of reference video URLs.

• Maximum 3 videos
• Total combined duration must not exceed 15 seconds
• Total combined size must not exceed 200 MB
• Each video must be between 720px and 2160px in resolution
• Can be combined with referenceImages
• Cannot be used with first_image or last_image

Legacy Compatibility

The legacy fields image (alias for first_image) and lastFrameImage (alias for last_image) are still supported for backwards compatibility.

If you pass both a legacy field and its new counterpart with conflicting values, the API will return a parameter conflict error. When both are present with the same value, first_image/last_image take precedence.

Response Fields

id · string

Unique task identifier. Save this value — you’ll use it to poll the Query Video Task endpoint.

object · string

Object type. Always "video".

created · integer

Unix timestamp of when the task was created.

model · string

The model name used for this task.

status · string

Task status at creation. Always "queued" on successful submission.

Lifecycle values:

• queued — task accepted and waiting in queue
• processing — model is actively generating
• completed — generation finished; video_url is available
• failed — generation failed; see error

error · string

Error message. null on successful submission.

Generation Modes

Mode	Fields Required	Notes
Text-to-video	model, prompt, duration	Optional: ratio, resolution
First/last frame	model, prompt, duration, first_image, last_image	Both frame fields required together
Image reference	model, prompt, duration, referenceImages	Up to 4 images, each < 20 MB
Video reference	model, prompt, duration, referenceVideos	Up to 3 videos, ≤15s total
Image + video	model, prompt, duration, referenceImages, referenceVideos	Combined constraints apply

Code Examples

cURL (Text-to-Video)

curl --request POST \
  --url https://zcbservice.aizfw.cn/kyyReactApiServer/v1/videos/videos \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "videos",
    "prompt": "A cute kitten playing on a sunlit meadow, cinematic lens",
    "duration": 5,
    "ratio": "16:9",
    "resolution": "720p"
  }'

Example Response

{
  "id": "video_1234567890",
  "object": "video",
  "created": 1774836724,
  "model": "videos",
  "status": "queued",
  "error": null
}

Next Steps

After receiving the task id, poll the Query Video Task endpoint to check status and retrieve video_url when generation completes.