Image generation API

Endpoints and authentication

Images live on the OpenAI-compatible entry point and use the same key as text models: only the path changes.

The key goes in Authorization: Bearer sk-your-tokenator-key. There is no separate image key — generations are drawn from the same key as tokens.

Your first request

curl https://api.tokenator.top/v1/images/generations \
  -H "Authorization: Bearer sk-your-tokenator-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "a white siamese cat on a windowsill",
    "n": 1,
    "size": "1024x1024"
  }'

{
  "created": 1755600000,
  "data": [
    { "url": "https://generated-static.tokenator.top/static/8f1c2a9b.png" }
  ]
}

The response carries exactly two fields: created, a unix timestamp, and data, one entry per requested image. An entry has either url or b64_json — the whole file in base64 when the model answered inline. The proxy adds nothing else.

The link points at a separate file host — generated-static.tokenator.top, path /static/. It is neither the API address nor the provider's temporary URL: the file was downloaded and rehosted on our side, is served with Cache-Control: immutable and lives as long as it sits on the service's disk.

Which models draw

Request parameters

Editing and upscaling

An existing image can be sent to /v1/images/edits — to upscale, retouch or change a fragment. The image field is the source file, prompt describes the edit, and an optional mask limits the area. The result is charged against the same quota as a plain generation.

curl https://api.tokenator.top/v1/images/edits \
  -H "Authorization: Bearer sk-your-tokenator-key" \
  -F model="gpt-image-2" \
  -F image="@input.png" \
  -F prompt="upscale to 4k, sharpen the details" \
  -F n=1 \
  -F size="2048x2048"

SDKs and ready-made examples

The endpoint is compatible with the OpenAI SDKs: only the base URL and the key change.

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-tokenator-key",
    base_url="https://api.tokenator.top/v1",
)

result = client.images.generate(
    model="gpt-image-2",
    prompt="a white siamese cat",
    n=1,
    size="1024x1024",
)

print(result.data[0].url)

import OpenAI from "openai"

const client = new OpenAI({
  apiKey: "sk-your-tokenator-key",
  baseURL: "https://api.tokenator.top/v1",
})

const result = await client.images.generate({
  model: "gpt-image-2",
  prompt: "a white siamese cat",
  n: 1,
  size: "1024x1024",
})

console.log(result.data[0].url)

Ready-made snippets with your own key and address are in the key dashboard, on the SDK / API tab.

How images are billed

Images do not spend tokens. A key carries a separate generation counter, and one generation is one image at the ordinary size. On some models 2048x2048 costs more: the multiplier is set by the operator and shown on the model page. The final charge is images × model multiplier, rounded up.

Generations are claimed before the provider is contacted and returned if no image arrives. A key with token_limit: -1 is a generation-only key with no access to text models.

curl https://api.tokenator.top/v1/tokens \
  -H "Authorization: Bearer sk-your-tokenator-key"

{
  "name": "my-key",
  "image_limit": 100,
  "image_used": 12,
  "image_remaining": 88
}

Limits

• Up to 10 images per request (n).
• Sizes: 1024x1024, 1024x1536, 1536x1024, 2048x2048.
• Concurrent generations per key: 1; beyond that a 429 with a busy_for_seconds field.
• Finished files are served from our address and do not depend on the provider's temporary links.

Errors

Trying it without code

The key dashboard has a Studio: the prompt, the size and the model are picked in the interface and finished images land in a gallery. It is the same endpoint and the same generation counter.

FAQ

Read next