AI-Now Image and Video Generation API Documentation

Table of Contents

1. Introduction
2. Authentication
3. API Endpoints
   • Ping
   • Generate
4. Models
5. Styles
6. Error Handling
7. Rate Limiting and Tokens
8. Examples

Introduction

This API provides access to various AI models for generating music, images, and videos based on text prompts. It supports multiple models with different capabilities and token costs.

Authentication

Authentication is required for all API requests. Use Bearer token authentication by including an API key in the Authorization header:

Authorization: Bearer YOUR_API_KEY

To obtain an API key, contact [email protected].

API Endpoints

Ping

Endpoint: GET /api/ping

Use this endpoint to check if the API is operational.

Response:

pong

Generate

Endpoint: POST /api/generate

This endpoint generates music, images, or videos based on the specified model and prompt.

Request Body:

{
  "model": "string",
  "prompt": "string",
  "style": "string" (optional)
}

• model: The AI model to use (e.g., "suno", "luma", "pika", "run")
• prompt: The text prompt for generation
• style: Optional style parameter for some models

Response:

The response is a server-sent event stream. Events include:

• Processing updates
• Error messages
• Final result with generated content URL

Models

1. Suno (Music Generation)

   • Token Cost: 10
   • Generation Time: 1-3 minutes

2. Luma (Video Generation)

   • Token Cost: 30
   • Generation Time: Variable, may fail during peak hours

3. Pika (Video Generation)

   • Token Cost: 10
   • Generation Time: 1-3 minutes

4. Run (Video Generation)

   • Token Cost: 30
   • Generation Time: 1-3 minutes

Styles

Pika API Styles

• Anime (value: '1')
• Moody (value: '2')
• 3D (value: '3')
• Watercolor (value: '4')
• Natural (value: '5')
• Claymation (value: '6')
• Black & White (value: '7')

Run API Styles

• Abandoned (value: 'abandoned')
• Abstract Sculpture (value: 'abstract_sculpture')
• Anime (value: 'anime')
• Cartoon (value: 'cartoon')
• Cine Lens (value: 'cine_lens')
• Claymation (value: 'claymation')
• Concept Art (value: 'concept_art')
• Digital Art (value: 'digital_art')
• Forestpunk (value: 'forestpunk')
• Frost (value: 'frost')
• Graphic Novel (value: 'graphic_novel')
• Graphite (value: 'graphite')
• Isometric 3D (value: 'isometric_3d')
• Low Poly 3D (value: 'low_poly_3d')
• Marker Drawing (value: 'marker_drawing')
• Moody Film (value: 'moody_film')
• Pixel Art (value: 'pixel_art')
• Retro (value: 'retro')
• Sci-fi Art (value: 'sci_fi_art')
• Stickers (value: 'stickers')
• Storyboard (value: 'storyboard')
• Thriller (value: 'thriller')
• 3D Cartoon (value: '3d_cartoon')
• 3D Render (value: '3d_render')
• 80s Vaporwave (value: '80s_vaporwave')

Error Handling

The API returns appropriate HTTP status codes and error messages:

• 401: Unauthorized (missing or invalid API key)
• 418: Insufficient tokens
• 500: Internal Server Error

Error responses are in JSON format:

{
  "status": "Error",
  "message": "Error description"
}

Rate Limiting and Tokens

Each API key has a token balance. Tokens are deducted for each generation request based on the model used. Ensure your account has sufficient tokens before making requests.

Examples

Generate a song using Suno

Request:

curl -X POST https://api.ai-now.space/api/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "suno",
    "prompt": "A upbeat pop song about summer"
  }'

Response:

The response is a server-sent event stream. Here's an example of the events you might receive:

data: {"status": "Processing started"}

data: {"status": "Processing"}

data: {"status": "Cookie generated"}

data: {"status": "Processing"}

data: {"status": "Completed", "url": "https://dl.ai-now.space/summer_pop_song_12345.mp3"}

Generate a video using Pika with a specific style

Request:

curl -X POST https://api.ai-now.space/api/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "pika",
    "prompt": "A serene lake surrounded by mountains",
    "style": "4"
  }'

Response:

data: {"status": "Processing started"}

data: {"status": "Processing"}

data: {"status": "Cookie generated"}

data: {"status": "Processing"}

data: {"status": "Processing"}

data: {"status": "Completed", "url": "https://dl.ai-now.space/serene_lake_mountains_67890.mp4"}

In these examples, the final event contains a url field with the download link for the generated content. The URL structure is:

https://dl.ai-now.space/<model>/<generated_file_name>.<extension>

Where:

• <model> is the name of the model used (e.g., "suno", "pika", "luma", "run")
• <generated_file_name> is a unique identifier for the generated content
• <extension> is the appropriate file extension for the content type (e.g., ".mp3" for audio, ".mp4" for video)

Note that the actual file names and identifiers will vary with each generation.