Skip to main content

​
Base URL

https://kinovi.ai/api/v1
All endpoints share the same base URL. Two endpoints are available:
EndpointMethodDescription
/jobs/createTaskPOSTCreate a generation task (video or image)
/jobs/recordInfoGETQuery task status and retrieve results

​
Authentication

All requests require a Bearer token in the Authorization header: 
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json
See Authentication for setup, security best practices, and rate limits.

​
Endpoints

Text-to-Video

Seedance 2.0 video from text

Image-to-Video

Seedance 2.0 video from image

Omni Reference

Visual references with @N syntax

Midjourney

Photorealistic and anime images

NanoBanana 2

Text-to-image with references

Get Task Status

Poll progress and retrieve results

​
Request Flow

All generation tasks are asynchronous. The flow is the same for every model:
1

Create Task

POST /jobs/createTask with your model and parameters. Returns a taskId.
2

Poll or Wait for Callback

Either poll GET /jobs/recordInfo?taskId=... every 3–5 seconds, or provide a callBackUrl to receive a webhook when complete.
3

Get Result

On success, the output array contains URLs to your generated videos or images.

​
Response Format

Create task returns a task ID: 
{
  "taskId": "task_clxxxxxx"
}
Query task returns the full result: 
{
  "taskId": "task_clxxxxxx",
  "model": "seedance-20",
  "status": "success",
  "creditsUsed": 200,
  "output": [
    {
      "url": "https://static.kinovi.ai/generated/video.mp4",
      "width": 1280,
      "height": 720
    }
  ],
  "error": null,
  "createTime": 1709900000000,
  "completeTime": 1709900120000
}
Error responses return a message with the appropriate HTTP status code: 
{
  "message": "Invalid request parameters"
}

​
Error Codes

HTTP StatusMeaningCommon Causes
200SuccessRequest completed
400Bad RequestInvalid parameters
401UnauthorizedInvalid or missing API key
402Payment RequiredInsufficient credits
403ForbiddenAPI access not enabled
404Not FoundTask doesn’t exist
429Too Many RequestsConcurrency limit reached
500Server ErrorInternal error (retry)
See Error Handling for detailed error codes, retry logic, and refund policy.

​
Supported Models

​
Video Models

ModelModel IDPricingOutput
Seedance 2.0 Proseedance-2040 cr/sec (720p)4–15s video
Seedance 2.0 Fastseedance2-fast28 cr/sec (720p)4–15s video, ~2x faster
Seedance 2.0 supports three generation modes: Text-to-Video, Image-to-Video, and Omni Reference.

​
Image Models

ModelModel IDPricingOutput
Midjourney V7midjourney-v712 credits ($0.06)4 images per run
Midjourney Niji V7midjourney-v7-niji12 credits ($0.06)4 images per run
NanoBanana 2nanobanana215–25 credits (0.075–0.075–0.125)1 image per run
See Midjourney and NanoBanana 2 for full parameters and examples.

​
Support