Image Generation

Overview

Token101 exposes an OpenAI Images-compatible API (/api/v1/images/generations). The route is available for OpenAI SDKs and raw HTTP clients, while live model availability is controlled by the model catalog. If an image model is temporarily paused, the request returns model_not_enabled instead of charging credits.

Endpoint

POST https://token.ppthub.shop/api/v1/images/generations

Authentication

Include your Token101 API key in the Authorization header:

Authorization: Bearer sk-your-token101-api-key

Quick Example

curl

curl https://token.ppthub.shop/api/v1/images/generations \
  --request POST \
  --header "Content-Type: application/json" \
  --header "Authorization: Bearer $TOKEN101_API_KEY" \
  --data '{
    "model": "gpt-image-1",
    "prompt": "A watercolor painting of a mountain lake at sunset",
    "n": 1,
    "size": "1024x1024"
  }'

Python (OpenAI SDK)

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["TOKEN101_API_KEY"],
    base_url="https://token.ppthub.shop/api/v1",
)

response = client.images.generate(
    model="gpt-image-1",
    prompt="A watercolor painting of a mountain lake at sunset",
    n=1,
    size="1024x1024",
)

image = response.data[0]
# Depending on the model, either `url` or `b64_json` is populated — check which one.
print(image.url or image.b64_json)

Node.js / TypeScript (OpenAI SDK)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.TOKEN101_API_KEY!,
  baseURL: 'https://token.ppthub.shop/api/v1',
});

const response = await client.images.generate({
  model: 'gpt-image-1',
  prompt: 'A watercolor painting of a mountain lake at sunset',
  n: 1,
  size: '1024x1024',
});

// Depending on the model, either `url` or `b64_json` is populated — check which one.
console.log(response.data[0]?.url ?? response.data[0]?.b64_json);

Request Fields

Field	Type	Description
`model`	string	Required. Image-capable model ID. Check Supported Models or `GET /api/v1/public/models` before dispatching.
`prompt`	string	Required. Text description of the desired image
`n`	integer	Optional. Number of images to generate (1–10, default: 1)
`size`	string	Optional. One of `auto`, `1024x1024`, `1024x1536`, `1536x1024`
`quality`	string	Optional. One of `auto`, `low`, `medium`, `high`
`response_format`	string	Optional. `url` or `b64_json`. Depending on the model, the response may only contain `b64_json` data — inspect the actual response shape rather than assuming a `url` is present.
`style`	string	Optional. Common values include `vivid` and `natural`

Response Format

A successful response returns a JSON object with the generated image data:

{
  "created": 1715659876,
  "data": [
    {
      "url": "https://..."
    }
  ]
}

When response_format is b64_json, the data array contains b64_json fields instead of URLs:

{
  "created": 1715659876,
  "data": [
    {
      "b64_json": "..."
    }
  ]
}

Model Availability

The image route and image models are managed separately. The route may be healthy even when a specific image model is paused for upstream stability. Use GET /api/v1/public/models and check each model's outputModalities and operationalStatus before enabling it in your client.

When a model is paused, Token101 returns an API error with reason: "model_not_enabled". That means the request format and API key were accepted, but live routing for that model is currently disabled. No successful image charge should be created for that failed request.

Image generation is priced per image, not per token. See Billing & Credits for current pricing and the control panel for the latest status.

Pricing

Image generation uses per-image pricing instead of per-token billing:

Item	Pricing behavior
Image generation	Charged per successfully generated image

The cost is deducted from your credits when the image is generated. Exact unit price, size/quality support, and model availability are shown in the live model catalog and billing pages.

Limitations

Non-streaming: Image generation does not support streaming. Requests block until the image is ready (typically 10–30 seconds).
Model support: Image-capable models can be temporarily paused. Check operationalStatus before sending production traffic.
Content policy: Prompts that violate OpenAI's content policy will be rejected by the upstream provider.
Rate limits: Image generation counts against your standard RPM limit but uses a separate concurrency budget due to longer processing times.

Troubleshooting

`401 Unauthorized`

Check your API key. See Authentication for details.

`400 Invalid model` or `model_not_enabled`

Confirm that the model exists and is currently enabled in GET /api/v1/public/models. model_not_enabled means the image route is reachable but live routing for that model is paused.

`429 Too Many Requests`

You've hit the rate limit. Image generation requests take longer than text requests, so concurrent limits may be reached sooner. See Rate Limits.