Banner generated by Reve AI

An unofficial Node.js SDK for interacting with Reve AI's image generation service

This SDK provides a programmatic interface for Reve AI's image generation capabilities, allowing developers to integrate Reve's new SOTA model in their applications before they release their own official API/SDKs.

Note: This is an unofficial SDK and is not affiliated with or endorsed by Reve AI. Use at your own risk.

Installation

npm install reve-sdk

Usage

import { ReveAI } from 'reve-sdk';

// Create a new instance of the Reve AI SDK with your authentication
const reveAI = new ReveAI({
  auth: {
    authorization: 'Bearer your-jwt-token-here', // Auth token from browser
    cookie: 'your-cookie-value-here', // Cookie value from browser
  },
  // Optional: specify project ID directly if you know it
  projectId: 'your-project-id-here', // Can be found in network requests of preview.reve.art after /api/project/
  // Optional: enable verbose logging for debugging
  verbose: true, // Set to true to see detailed request/response logs
  // Optional: add custom headers to every request
  customHeaders: { 'x-custom-header': 'value' },
  // Optional configuration
  timeout: 60000, // 60 seconds
  pollingInterval: 2000, // Poll every 2 seconds
  maxPollingAttempts: 60, // Maximum number of polling attempts
});

// Generate an image
async function generateImage() {
  try {
    // Generate an image with a prompt
    const result = await reveAI.generateImage({
      prompt: 'A beautiful landscape with mountains and a lake',
      negativePrompt: 'blur, low quality',
      width: 1024,
      height: 768,
      batchSize: 1, // How many images to generate
      seed: -1, // -1 for random seed
      model: 'text2image_v1/prod/20250325-2246', // Default model
      enhancePrompt: true, // Whether to enhance the prompt automatically
    });
    
    // Access generated image URLs
    console.log('Images generated:', result.imageUrls);
    console.log('Seed used:', result.seed);
    console.log('Completed at:', result.completedAt);
    console.log('Prompt:', result.prompt);
    console.log('Enhanced Prompt:', result.enhancedPrompt);
    
    // Save the images to files
    result.imageUrls.forEach((base64Data, index) => {
      // Extract the base64 data (removing the data URL prefix if present)
      const base64Image = base64Data.includes('base64,') 
        ? base64Data.split('base64,')[1] 
        : base64Data;
      
      // Convert base64 to binary
      const imageBuffer = Buffer.from(base64Image, 'base64');
      
      // Save to file
      const fs = require('fs');
      fs.writeFileSync(`image_${index}.webp`, imageBuffer);
    });
    
  } catch (error) {
    console.error('Error generating image:', error);
  }
}

generateImage();

How to get authentication values

To get the authorization token and cookie values:

1. Log in to the Reve AI website (https://preview.reve.art)
2. Open browser developer tools (F12) and go to the Network tab
3. Make an API request on the site (e.g., start a generation)
4. Find the request in the Network tab
5. Copy the "Authorization" header value from the request headers
6. Copy the "Cookie" header value from the request headers
7. Copy the project ID from the request URL (format: /api/project/{projectId}/generation)

Debugging

If you encounter issues with the SDK, you can enable verbose logging by setting the verbose option to true:

const reveAI = new ReveAI({
  // ...other options
  verbose: true
});

This will log detailed information about:

• All HTTP requests (URL, method, headers, request body)
• All HTTP responses (status code, headers, response body)
• Detailed error information with request/response context

The verbose mode is particularly helpful for troubleshooting authentication, API endpoint, or project ID issues.

Support Discord

If you need help using the SDK feel free to join my support Discord!

API Reference

ReveAI Class

Constructor

const reveAI = new ReveAI(options);

Options:

Parameter	Type	Required	Default	Description
auth	object	Yes	-	Authentication credentials
auth.authorization	string	Yes	-	Authorization header with Bearer token
auth.cookie	string	Yes	-	Cookie value for authentication
projectId	string	No	-	Project ID to use for generations. If not provided, the SDK will try to auto-detect
baseUrl	string	No	'https://preview.reve.art'	Reve AI API base URL
timeout	number	No	30000	Request timeout in milliseconds
maxPollingAttempts	number	No	60	Maximum number of status check attempts
pollingInterval	number	No	2000	Interval between status checks in milliseconds
verbose	boolean	No	false	Enable verbose logging for debugging
customHeaders	object	No	{}	Custom headers to add to every request

Methods

generateImage(options)

Generates an image based on the provided options.

const result = await reveAI.generateImage(options);

Options:

Parameter	Type	Required	Default	Description
prompt	string	Yes	-	Text prompt describing the image to generate
negativePrompt	string	No	''	Text prompt describing what to exclude from the image
width	number	No	1024	Image width in pixels (384-1024, divisible by 8)
height	number	No	1024	Image height in pixels (384-1024, divisible by 8)
batchSize	number	No	1	Number of images to generate (1-4)
seed	number	No	-1	Random seed (-1 for random)
model	string	No	'text2image_v1/prod/20250325-2246'	Model name to use for generation
enhancePrompt	boolean	No	true	Whether to enhance the prompt automatically using Reve AI's prompt enhancement model

Return value:

{
  imageUrls: string[]; // Array of URLs to the generated images
  seed: number; // The seed that was used for generation
  completedAt: Date; // Timestamp when the generation was completed
  prompt: string; // The original prompt
  enhancedPrompt?: string; // The enhanced prompt used for generation (if enhancePrompt was enabled)
  enhancedPrompts?: string[]; // Array of all enhanced prompts used when generating multiple images
  negativePrompt?: string; // The negative prompt, if provided
}

Prompt Enhancement

When enhancePrompt is set to true (default), the SDK will send your prompt to Reve AI's prompt enhancement model before generating the image. This model expands your prompt with additional details to improve the quality and consistency of the generated images.

For example, a simple prompt like "a superhero advertising Red Bull" might be enhanced to include details about composition, lighting, style, and more specific visual elements.

The enhanced prompt is available in the enhancedPrompt property of the result object.

When generating multiple images in a batch (using batchSize > 1), the SDK will:

1. Request multiple prompt variants from the enhancement model (equal to your batch size)
2. Use a different enhanced prompt for each image in the batch
3. Return all the enhanced prompts used in the enhancedPrompts array

This ensures maximum variation between the generated images in a batch, as each one will use a uniquely enhanced version of your original prompt.

// Generate 4 images with different enhanced prompts
const result = await reveAI.generateImage({
  prompt: "a superhero advertising Red Bull",
  batchSize: 4
});

// Access all enhanced prompts used
console.log(result.enhancedPrompts);

Error Handling

The SDK throws ReveAIError instances for any errors that occur. The error includes a type property to help identify the specific error category:

try {
  await reveAI.generateImage({ prompt: 'A beautiful sunset' });
} catch (error) {
  if (error.name === 'ReveAIError') {
    console.error(`Error type: ${error.type}`);
    console.error(`Message: ${error.message}`);
    if (error.statusCode) {
      console.error(`Status code: ${error.statusCode}`);
    }
  }
}

Error types:

• AUTHENTICATION_ERROR - Issues with authentication
• API_ERROR - Server returned an error response
• REQUEST_ERROR - Network or request formation problems
• TIMEOUT_ERROR - Request timed out
• GENERATION_ERROR - Image generation failed
• POLLING_ERROR - Exceeded maximum polling attempts
• UNEXPECTED_RESPONSE - Server response didn't match expected format
• UNKNOWN_ERROR - Other unexpected errors

Development

Prerequisites

• Node.js (20.x or later)
• npm or yarn

Setup

1. Clone the repository
2. Install dependencies:

   npm install

3. Build the project:

   npm run build

Testing

Run tests:

npm test

Run tests in watch mode:

npm run test:watch

Linting and Formatting

Lint the code:

npm run lint

Format the code:

npm run format

Contributing

Feel free to submit any PRs, here's how to get going:

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a PR

Please make sure your code passes all tests and linting rules before submitting a PR. (thank you)