Kino API

Generate video with Seedance 2.0 through a simple asynchronous task API: create a task, poll for completion, download your video from a CDN URL. All requests are plain HTTPS + JSON.

Authentication

All API calls require a Bearer token. Create tokens in Console → Tokens; each token has its own balance scope and can be revoked at any time.

Authorization: Bearer sk-YOUR_TOKEN

Quickstart

# Create a task
curl -X POST https://api.kino-api.com/v1/video/generations \
  -H "Authorization: Bearer sk-YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-mini",
    "prompt": "Wind turbines on a green hillside at dawn, cinematic",
    "duration": 5,
    "metadata": { "resolution": "480p", "ratio": "16:9" }
  }'

# → {"task_id": "task_3ycw...", "status": "queued"}

# Poll every 3–5 seconds
curl https://api.kino-api.com/v1/video/generations/task_3ycw... \
  -H "Authorization: Bearer sk-YOUR_TOKEN"

# → {"data": {"status": "SUCCESS", "result_url": "https://.../video.mp4"}}

Models

Create Video Task

Request body

Response

{
  "task_id": "task_3ycweCkT2zWiAQwlzTSy1Y3UlV5dYFDs",
  "status": "queued",
  "model": "seedance-2.0-mini",
  "created_at": 1785989680
}

Poll Task Status

Poll every 3–5 seconds. Typical generation takes 30–90 seconds.

{
  "data": {
    "status": "SUCCESS",
    "result_url": "https://media.uptoken.cc/v/ut-Vc3cN2bcNVwq.mp4",
    "quota": 55702,
    "submit_time": 1785990148,
    "finish_time": 1785990233
  }
}

Generate Image (Synchronous)

Unlike video, image generation is synchronous: one request returns the finished image in about 30 seconds — no polling needed. OpenAI-compatible request and response shape.

Request body

Example

curl -X POST https://api.kino-api.com/v1/images/generations \
  -H "Authorization: Bearer sk-YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedream-5.0-lite",
    "prompt": "A tiny paper boat floating in a puddle after rain, macro photography"
  }'

# → {"created": 1785998459, "data": [{"url": "https://.../image.jpg"}]}

Pricing

Parameter Reference (metadata)

Image / Video Inputs

Image-to-video

{
  "model": "seedance-2.0-pro",
  "prompt": "The camera slowly pushes in as waves begin to move",
  "images": ["https://example.com/frame.jpg"],
  "metadata": { "resolution": "1080p" }
}

Video-reference mode

Pass a reference video through metadata.content. Video-reference tasks are billed at the higher tier (see Pricing).

{
  "model": "seedance-2.0-pro",
  "prompt": "Restyle this clip into cyberpunk anime",
  "metadata": {
    "content": [{ "type": "video_url",
                   "video_url": { "url": "https://example.com/ref.mp4" } }]
  }
}

Billing & Refunds

Billing is purely usage-based: each successful video is charged by resolution × duration at the per-second rates on the pricing table. Charges are computed from upstream usage, so you always pay exactly the listed rate — no per-call fees, no rounding up.

Failed tasks are refunded automatically and in full, the moment the upstream reports failure. Your console logs show the exact charge for every request.

Errors

Rate Limits & Best Practices

Best practices: poll with backoff, don't retry identical prompts immediately, download result_url promptly, use separate tokens per environment.

SDK Examples

Python

import requests, time

BASE, TOKEN = "https://api.kino-api.com", "sk-YOUR_TOKEN"
H = {"Authorization": f"Bearer {TOKEN}"}

task = requests.post(f"{BASE}/v1/video/generations", headers=H, json={
    "model": "seedance-2.0-mini",
    "prompt": "Wind turbines on a green hillside at dawn",
    "duration": 5,
    "metadata": {"resolution": "480p", "ratio": "16:9"},
}).json()

while True:
    time.sleep(4)
    r = requests.get(f"{BASE}/v1/video/generations/{task['task_id']}", headers=H).json()["data"]
    if r["status"] == "SUCCESS":
        print("Video:", r["result_url"]); break
    if r["status"] == "FAILURE":
        raise RuntimeError(r["fail_reason"])

Node.js

const BASE = "https://api.kino-api.com", TOKEN = "sk-YOUR_TOKEN";
const H = { Authorization: `Bearer ${TOKEN}`, "Content-Type": "application/json" };

const t = await (await fetch(`${BASE}/v1/video/generations`, {
  method: "POST", headers: H,
  body: JSON.stringify({ model: "seedance-2.0-mini",
    prompt: "Wind turbines on a green hillside at dawn",
    duration: 5, metadata: { resolution: "480p", ratio: "16:9" } })
})).json();

while (true) {
  await new Promise(r => setTimeout(r, 4000));
  const { data } = await (await fetch(`${BASE}/v1/video/generations/${t.task_id}`, { headers: H })).json();
  if (data.status === "SUCCESS") { console.log("Video:", data.result_url); break; }
  if (data.status === "FAILURE") throw new Error(data.fail_reason);
}

Questions? Open a ticket via the console or check the FAQ.