Docs · EasyToken

Quickstart

The API lives at https://api.easytoken.si/v1. Every model uses the same key. Video, image and audio models run as tasks: you create one, then receive the result by webhook or by polling. Chat models answer in the same request.

Create an account and copy your API key from the dashboard.
Send your first task with the request below.
Open the result.url when the task finishes.

first-task.sh

curl https://api.easytoken.si/v1/tasks \
  -H "Authorization: Bearer $EASYTOKEN_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.5",
    "input": { "prompt": "Product spin on a marble table" }
  }'

Authentication

Send your key as a bearer token in the Authorization header. Keys start with et_live_. Keep them on your server; never ship them in a browser or mobile app.

You can create several keys, each with its own daily spend cap, from Dashboard → API keys.

Media tasks

POST /v1/tasks creates a video, image or audio job and returns at once with a task ID.

Field	Type	Description
`model`	string	Model ID, e.g. `veo-3.1`. See Model IDs.
`input`	object	Model-specific input. Every model accepts `prompt` (ElevenLabs also takes `text` and `voice`). Video and image models accept `aspect_ratio` and `image_urls` (https links) for image-to-video or editing. Suno accepts `style`, `lyrics`, `title` and `instrumental`.
`callback_url`	string	Optional public https URL. We POST the finished task here.
`metadata`	object	Optional. Returned unchanged, useful for your own IDs.

response · 202

{
  "id": "task_8f2kq1x",
  "model": "seedance-2.5",
  "status": "queued",
  "cost_usd": 1.95,
  "created_at": "2026-10-07T12:00:00Z"
}

Get a task

GET /v1/tasks/{id} returns the current state. Output links stay valid for 14 days, so copy files you want to keep. status moves through queued → running → succeeded or failed. Poll every 5 to 10 seconds, or use a webhook instead.

succeeded

{
  "id": "task_8f2kq1x",
  "status": "succeeded",
  "result": { "url": "https://…/output.mp4", "urls": ["https://…/output.mp4"], "type": "video" },
  "cost_usd": 1.95
}

Chat completions

POST /v1/chat/completions follows the OpenAI format, including streaming, tools and JSON mode. Point any OpenAI SDK at our base URL.

chat.js

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.easytoken.si/v1",
  apiKey: process.env.EASYTOKEN_KEY,
});

const res = await client.chat.completions.create({
  model: "claude-opus-5-5",
  messages: [{ role: "user", content: "Summarise this brief in 3 bullets" }],
});

Webhooks

When you pass callback_url, we send the same body as Get a task once the task finishes. Each request carries an X-EasyToken-Signature header: an HMAC-SHA256 of the raw body with your webhook secret. Reply with any 2xx within 10 seconds; we retry up to 5 times with backoff.

Model IDs

GET /v1/models returns this list as JSON with prices; GET /v1/balance returns your credit balance.

Model ID	Type	Endpoint	Price

Errors

Status	Code	What to do
401	`invalid_api_key`	Check the key and that it hasn't been revoked.
402	`insufficient_credits`	Top up in the dashboard, or raise the key's spend cap.
404	`model_not_found`	Use an ID from the Model IDs table.
422	`invalid_input`	The message names the field that failed.
429	`rate_limited`	Wait for the `Retry-After` seconds, then resend.
5xx	`upstream_error`	Safe to retry. Failed tasks are never billed.

Rate limits

New accounts can run 10 media tasks at once per model and 60 chat requests per minute. Limits rise automatically with usage; contact support if you need more for a launch.

Billing

Usage is paid from prepaid credits. Each task's price is fixed when it is created and shown in cost_usd. Tasks that fail, time out or are blocked by a safety filter are refunded automatically.