🎨 HunyuanImage 2.1 Pipeline

HunyuanImage-2.1: An Efficient Diffusion Model for High-Resolution (2K) Text-to-Image Generation​

This app provides three main functionalities:

  1. Text-to-Image Generation: Generate high-quality images from text prompts
  2. Prompt Enhancement: Improve your prompts using MLLM reprompting
  3. Image Refinement: Enhance existing images with the refiner model

Generation Settings

Model: HunyuanImage v2.1 (Non-distilled)
Aspect Ratio

Select the aspect ratio for image generation
10 100
1 10

Enhance prompt automatically

Apply refiner after generation

Generated Image

📝 Usage Tips

Text-to-Image Generation:

  • Use descriptive prompts with specific details
  • Adjust guidance scale: higher values follow prompts more closely
  • More inference steps generally produce better quality
  • Enable reprompt for automatic prompt enhancement
  • Enable refiner for additional quality improvement

Prompt Enhancement:

  • Enter your basic prompt idea
  • The AI will enhance it with better structure and details
  • Enhanced prompts often produce better results

Image Refinement:

  • Upload any image you want to improve
  • Describe what improvements you want in the refinement prompt
  • The refiner will enhance details and quality
  • Works best with images generated by HunyuanImage