media

Make an image or video with AI

Paid in the workspace's credits at a fixed price per model. First call with estimate_only: true (free) for the exact price, then again with max_credits set to it. An image is usually done within wait_seconds; a video takes one to five minutes: GET the generation by its id. A failed generation gives its credits back.

POST/v1/sites/{site_id}/media/generations

Request

curl --request POST \
  --url 'https://api.atelier201.com/v1/sites/<site_id>/media/generations' \
  --header 'Authorization: <api-key>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "gpt-image",
  "prompt": "<string>"
}'

import json
import requests

url = "https://api.atelier201.com/v1/sites/<site_id>/media/generations"
headers = {
  "Authorization": "<api-key>",
  "Content-Type": "application/json"
}
payload = json.loads("{\"model\":\"gpt-image\",\"prompt\":\"<string>\"}")

response = requests.request("POST", url, headers=headers, json=payload)
print(response.json())

const response = await fetch("https://api.atelier201.com/v1/sites/<site_id>/media/generations", {
  method: "POST",
  headers: {
    "Authorization": "<api-key>",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    "model": "gpt-image",
    "prompt": "<string>"
  }),
});

const data = await response.json();
console.log(data);

<?php
$client = curl_init();

curl_setopt($client, CURLOPT_URL, "https://api.atelier201.com/v1/sites/<site_id>/media/generations");
curl_setopt($client, CURLOPT_CUSTOMREQUEST, "POST");
curl_setopt($client, CURLOPT_HTTPHEADER, [
    "Authorization: <api-key>",
    "Content-Type: application/json",
]);
curl_setopt($client, CURLOPT_POSTFIELDS, "{\"model\":\"gpt-image\",\"prompt\":\"<string>\"}");
curl_setopt($client, CURLOPT_RETURNTRANSFER, true);

$response = curl_exec($client);
curl_close($client);
echo $response;

package main

import (
  "bytes"
  "fmt"
  "io"
  "net/http"
)

func main() {
  req, _ := http.NewRequest("POST", "https://api.atelier201.com/v1/sites/<site_id>/media/generations", bytes.NewBuffer([]byte("{\"model\":\"gpt-image\",\"prompt\":\"<string>\"}")))
  req.Header.Set("Authorization", "<api-key>")
  req.Header.Set("Content-Type", "application/json")
  res, _ := http.DefaultClient.Do(req)
  defer res.Body.Close()
  data, _ := io.ReadAll(res.Body)
  fmt.Println(string(data))
}

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;

var request = HttpRequest.newBuilder(URI.create("https://api.atelier201.com/v1/sites/<site_id>/media/generations"))
    .header("Authorization", "<api-key>")
    .header("Content-Type", "application/json")
    .method("POST", HttpRequest.BodyPublishers.ofString("{\"model\":\"gpt-image\",\"prompt\":\"<string>\"}"))
    .build();

var response = HttpClient.newHttpClient().send(
    request, HttpResponse.BodyHandlers.ofString());
System.out.println(response.body());

require 'net/http'
require 'json'

uri = URI("https://api.atelier201.com/v1/sites/<site_id>/media/generations")
request = Net::HTTP::Post.new(uri)
request["Authorization"] = "<api-key>"
request["Content-Type"] = "application/json"
request.body = "{\"model\":\"gpt-image\",\"prompt\":\"<string>\"}"

response = Net::HTTP.start(uri.hostname, uri.port, use_ssl: uri.scheme == "https") do |http|
  http.request(request)
end

puts response.body

application/json

{
  "data": {
    "id": "<string>",
    "status": "running",
    "media": {
      "path": "<string>",
      "url_path": "<string>",
      "mime": "<string>",
      "size": 123,
      "alt": null,
      "credit": null,
      "keep_original": true
    },
    "error": null,
    "credits": 123,
    "credits_refunded": 123,
    "estimated_credits": 123,
    "balance": 123.45,
    "next_action": null
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

{
  "error": {
    "type": "<string>",
    "code": "<string>",
    "message": "<string>",
    "param": "<string>",
    "docs_url": "<string>",
    "action_url": "<string>",
    "details": {}
  },
  "request_id": "<string>"
}

Requires media:generate, a permission of its own: a credential that can edit sites cannot spend credits unless it was given this one too. Spends the workspace’s credits.

Call it with estimate_only: true. It costs nothing and answers estimated_credits (the exact price) and balance.
Call it again with max_credits set to that price. Without max_credits the call is refused. A price above it is refused with nothing charged: invalid_parameter on max_credits, with details.reason price_above_max_credits, details.estimated_credits and details.max_credits.

The credits are charged when the generation starts. If it fails, they are given back. An image is usually done within wait_seconds (50 by default) and comes back with status: succeeded and its media. A video takes one to five minutes: the answer is status: running with an id; get the generation until it finishes.

The file lands in the site’s media like an upload, with credit set to the model that made it. Never present a generated image as a photograph of the business, its products, its people or its premises.

At most 10 generations a minute per workspace (429 rate_limited above that, with details.retry_after_seconds). Not enough credits: 402 insufficient_credits with details.needed and details.balance. A model that is busy fails the generation with its credits given back: try again in a minute.

Authorizations

Authorizationbearerin headerrequired

Path parameters

site_idstringrequired

The site's id, or its slug (the address before .atelier201.page).

Body

application/jsonrequired

modelenum<string>requiredone of: gpt-image, nano-banana, flux, p-image, wan, minimax, seedance, p-video

GET /v1/media/models lists each model, what it is best at and its prices in credits.

promptstringrequiredlength: 1–4000 chars

What to show: subject, setting, light, framing, style.

aspect_ratioenum<string>one of: 16:9, 1:1, 9:16, 4:3, 3:4, 3:2, 2:3

Width to height. Each model accepts the ratios list_media_models gives; default its first (16:9 for most).

qualityenum<string>one of: draft, standard, high

Images: draft, standard (default) or high.

durationintegerrange: 1–10

Videos: seconds.

resolutionenum<string>one of: 480p, 720p, 768p, 1080p, 2k

Videos: default the model's (720p, or 768p for minimax and p-video).

audioboolean

Videos: with sound (seedance only).

filenamestringlength: 1–100 chars

A name without extension.

altstringlength: ≤ 300 chars

estimate_onlyboolean

true: the exact price in credits and the balance. Free, nothing runs.

max_creditsintegerrange: ≤ 9007199254740991

Required to run: the most you agree to spend. Refused, and nothing charged, when the price is higher. Take it from an estimate_only call.

wait_secondsintegerrange: 0–55

How long to wait for the result. Default 50 for an image, 0 for a video.

Response

application/json

Make an image or video with AI

dataRecord<string, object>required

Show child attributes

data.idstring

The generation's id: GET it until status is succeeded or failed.

data.statusenum<string>one of: running, succeeded, failed

data.mediaobject | null

Show child attributes

data.media.pathstringrequired

data.media.url_pathstringrequired

data.media.mimestringrequired

data.media.sizeintegerrequiredrange: -9007199254740991–9007199254740991

data.media.altstringnullrequired

data.media.creditstringnullrequired

Who made it, where a licence asks for it: "Photo by Jane Doe on Pixabay", "Generated with Seedream 5 Lite (ByteDance)". Null for an upload.

data.media.keep_originalbooleanrequired

Stored exactly as uploaded, and never re-encoded by a build. Otherwise a picture over 200 KB is compressed when it arrives (2000 px at most, towards 150 KB, quality 75 or more, metadata removed).

data.errorstringnull

data.creditsinteger | null

What it cost.

data.credits_refundedinteger | null

Given back after a failure.

data.estimated_creditsintegerrange: -9007199254740991–9007199254740991

estimate_only: the exact price it would cost.

data.balancenumber

estimate_only: the workspace's credits now.

data.next_actionstringnullrequired

request_idstringrequired

Make an image or video with AI

Authorizations

Path parameters

Body

Response

Use these docs with your AI tools

Query these docs via MCP

Query these docs over HTTP

Read these docs as Markdown

Cookies