API Overview

External API authentication, models, chat completions, image generation, and credits.

Base URL

Use your site origin as the API base URL.

https://your-domain.com/api/v1

Authentication

Create an API key in Settings -> API Keys, then send it as a bearer token:

Authorization: Bearer sk_live_...

API keys are shown only once when created. The platform stores only the hashed value and a short prefix.

Available Models

List the currently enabled public models:

curl https://your-domain.com/api/v1/models \
  -H "Authorization: Bearer $YOUR_API_KEY"

Example response:

{
  "object": "list",
  "data": [
    {
      "id": "openai/gpt-5",
      "object": "model",
      "owned_by": "openrouter",
      "display_name": "GPT-5",
      "category": "chat",
      "capabilities": ["chat", "reasoning"],
      "credit_cost": 3
    }
  ]
}

Chat Completions

Endpoint:

POST /api/v1/chat/completions

Request:

curl https://your-domain.com/api/v1/chat/completions \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5",
    "messages": [
      { "role": "user", "content": "Write a haiku about APIs." }
    ]
  }'

Node.js example:

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.YOUR_API_KEY,
  baseURL: 'https://your-domain.com/api/v1',
});

const completion = await client.chat.completions.create({
  model: 'openai/gpt-5',
  messages: [{ role: 'user', content: 'Write a haiku about APIs.' }],
});

console.log(completion.choices[0]?.message?.content);

The chat endpoint enforces the platform model whitelist before routing the request upstream.

Image Generation

Endpoint:

POST /api/v1/images/generations

Request:

curl https://your-domain.com/api/v1/images/generations \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/nano-banana-pro",
    "prompt": "A cinematic photo of a robot reading API docs",
    "size": "1024x1024"
  }'

Response:

{
  "created": 1710000000,
  "data": [],
  "task_id": "local-task-id",
  "provider_task_id": "provider-task-id",
  "status": "pending",
  "model": "google/nano-banana-pro"
}

The first version returns task metadata immediately. Generated assets continue through the platform task pipeline.

Credits And Limits

Chat and image requests consume platform credits.
Credit cost is defined per model in the platform model registry.
The request is rejected before execution when the current user balance is insufficient.
API keys also carry a per-minute rate limit field for future enforcement.

Error Format

All /api/v1/* endpoints return a consistent error body:

{
  "error": {
    "message": "Model is not available",
    "type": "model_not_available",
    "code": "model_not_available"
  }
}

Common error codes:

invalid_api_key
api_disabled
model_required
model_not_available
messages_required
prompt_required
insufficient_credits
provider_not_configured
provider_error

Notes

Only models exposed by the platform model registry can be used.
API request logs are recorded with request ID, key, model, status, latency, and credit cost.
Videos and music are not part of the public API in this version.

API Overview

On this page