Introduction#

The CelebifyAI API lets you generate images, videos, upscale media, and perform face swaps programmatically. All endpoints follow REST conventions and return JSON.

Quick Start

curl https://celebifyai.com/api/v1/account \
  -H "Authorization: Bearer cel_your_api_key"

Every request must include your API key in the Authorization header as a Bearer token. All responses are JSON with appropriate HTTP status codes.

Authentication#

Authenticate by including your API key as a Bearer token in the Authorization header of every request.

Authorization: Bearer cel_your_api_key

API keys start with the prefix cel_. You can create and manage keys in the Developer Portal.

Rate Limits#

API access requires an active subscription (Base, Plus, or Pro). Rate limits are based on concurrent generations rather than requests per second. Polling status endpoints does not count toward the limit.

Errors#

When an error occurs the API returns a JSON object with an error field containing a machine-readable code and a human-readable message.

{
  "error": {
    "code": "insufficient_credits",
    "message": "You need 15 credits but only have 3 remaining.",
    "details": {
      "required": 15,
      "available": 3
    }
  }
}

Pagination#

List endpoints use cursor-based pagination. Pass limit and cursor as query parameters.

GET /api/v1/images?limit=20&cursor=clx123abc

Response format:

{
  "data": [ ... ],
  "has_more": true,
  "next_cursor": "clx456def"
}

When has_more is true, pass next_cursor as the cursor parameter in your next request to fetch the next page. The default limit is 20, maximum is 100.

Account#

Retrieve your current account status including credit balance, plan details, and concurrency usage.

Models#

List all available models. Optionally filter by type. Each model includes its ID, name, type, credit cost, and maximum quantity per request.

Image Models

Video Models — Text-to-Video

Video Models — Image-to-Video

Use the GET /api/v1/models endpoint for the complete, up-to-date list of models and their parameters.

Image Generation#

Generate images using any available image model. Submit a generation request, then poll for the result or use the list endpoint to retrieve your history.

Video Generation#

Generate videos from text prompts or images. Video generation typically takes longer than image generation (30 seconds to several minutes depending on the model).

Face Swap#

Swap faces in images and videos. Provide a source face image and a target media file.

Upscaling#

Upscale images and videos to higher resolutions using AI-powered super-resolution.

Credit Costs#

Below is a complete reference of credit costs for all operations. Credits are deducted when a generation starts. If a generation fails due to a server error, credits are automatically refunded.

Image Models

Video Models

Other Operations