ai-image

A unified wrapper for image generation APIs (OpenAI and Replicate) with CLI support and programmatic access.

Features

• 🎨 Support for multiple image generation providers (OpenAI DALL-E, Replicate)
• 🖥️ Easy-to-use CLI interface
• 📦 Programmatic API for integration into your projects
• 🔐 Environment variable support for API keys
• 📁 Flexible output options (directory, filename)
• 🎯 TypeScript support with full type definitions

Installation

npm install image-gen-wrapper

Or install globally for CLI usage:

npm install -g image-gen-wrapper

Setup

API Keys

Create a .env file in your project root:

OPENAI_API_KEY=your_openai_api_key_here
REPLICATE_API_TOKEN=your_replicate_token_here

Get your API keys:

• OpenAI: https://platform.openai.com/api-keys
• Replicate: https://replicate.com/account/api-tokens

CLI Usage

Basic Usage

# Generate with OpenAI (default)
image-gen generate "A sunset over mountains"

# Generate with Replicate
image-gen generate "A cyberpunk city at night" --provider replicate

# Specify output file
image-gen generate "A cute robot" --output robot.png

# Generate multiple images
image-gen generate "Abstract art" --number 3

Advanced Options

# Use specific model
image-gen generate "Portrait of a wizard" --model dall-e-2

# Custom size and quality
image-gen generate "Ocean waves" --size 1792x1024 --quality hd

# Pass API key directly
image-gen generate "Forest path" --api-key sk-...

Available Commands

# Generate images
image-gen generate <prompt> [options]

# List available models
image-gen models

# Show setup instructions
image-gen setup

Options

• -p, --provider <provider> - Provider to use (openai or replicate)
• -k, --api-key <key> - API key (overrides environment variable)
• -o, --output <path> - Output file path
• -d, --output-dir <dir> - Output directory
• -m, --model <model> - Model to use
• -s, --size <size> - Image size (e.g., 1024x1024)
• -q, --quality <quality> - Image quality (standard or hd) - OpenAI only
• --style <style> - Image style (vivid or natural) - OpenAI only
• -n, --number <n> - Number of images to generate

Programmatic Usage

Basic Example

import { ImageGenerator } from "image-gen-wrapper";

// Create generator instance
const generator = new ImageGenerator({
  provider: "openai",
  outputDir: "./images",
});

// Generate image
const paths = await generator.generate({
  prompt: "A serene lake at dawn",
  size: "1024x1024",
  quality: "hd",
});

console.log("Images saved to:", paths);

Using Replicate

import { ImageGenerator } from "image-gen-wrapper";

const generator = new ImageGenerator({
  provider: "replicate",
  apiKey: "your-api-key", // Optional if using .env
  outputFilename: "my-image.png",
});

const paths = await generator.generate({
  prompt: "A futuristic space station",
  model:
    "stability-ai/sdxl:39ed52f2a78e934b3ba6e2a89f5b1c712de7dfea535525255b1aa35c5565e08b",
  n: 2,
});

Custom Output Handling

import { ImageGenerator } from "image-gen-wrapper";
import path from "path";

const generator = new ImageGenerator({
  provider: "openai",
  outputDir: path.join(__dirname, "output"),
  outputFilename: "custom-name.png",
});

// Generate multiple images with custom naming
const paths = await generator.generate({
  prompt: "Abstract geometric patterns",
  n: 3,
});
// Will save as: custom-name.png, custom-name_1.png, custom-name_2.png

Supported Models

OpenAI

• dall-e-3 (default) - Latest DALL-E model
• dall-e-2 - Previous generation

Replicate

• stability-ai/sdxl (default) - Stable Diffusion XL
• stability-ai/stable-diffusion - Stable Diffusion
• Any model from replicate.com in format owner/model:version

Error Handling

try {
  const paths = await generator.generate({
    prompt: "Mountain landscape",
  });
} catch (error) {
  console.error("Generation failed:", error.message);
}

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.