Skip to main content

ImageAgent

ImageAgent provides image generation and analysis capabilities.

Quick Start

import { createImageAgent } from 'praisonai';

const agent = createImageAgent({
  llm: 'openai/gpt-4o-mini'
});

// Generate an image
const images = await agent.generate({
  prompt: 'A sunset over mountains',
  size: '1024x1024'
});

// Analyze an image
const analysis = await agent.analyze({
  imageUrl: 'https://example.com/image.jpg',
  prompt: 'Describe this image'
});

Configuration

interface ImageAgentConfig {
  name?: string;
  llm?: string;
  imageModel?: string;  // Default: 'dall-e-3'
  verbose?: boolean;
}

Image Generation

interface ImageGenerationConfig {
  prompt: string;
  size?: '256x256' | '512x512' | '1024x1024' | '1792x1024' | '1024x1792';
  quality?: 'standard' | 'hd';
  style?: 'vivid' | 'natural';
  n?: number;  // Number of images
}

const images = await agent.generate({
  prompt: 'A futuristic city',
  size: '1024x1024',
  quality: 'hd',
  style: 'vivid'
});

Image Analysis

interface ImageAnalysisConfig {
  imageUrl: string;
  prompt?: string;
  detail?: 'low' | 'high' | 'auto';
}

const analysis = await agent.analyze({
  imageUrl: 'https://example.com/photo.jpg',
  prompt: 'What objects are in this image?',
  detail: 'high'
});

CLI Usage

# Generate image
praisonai-ts image generate "A sunset over mountains"
praisonai-ts image generate "A cat" --size 1024x1024 --quality hd

# Analyze image
praisonai-ts image analyze https://example.com/image.jpg
praisonai-ts image analyze https://example.com/image.jpg "What is this?"