Midjourney Prompt Generator

Midjourney Prompt Generator is a web application with no server component; user prompts and settings are not stored, and everything is generated locally in the browser. The core function is to generate Midjourney prompts from text input and various options. The provided parameter options are extremely rich:

• Medium: 18 options including Acrylic, Watercolor, Oil, Photography, Digital art, etc.
• Lighting: 51 options including Cinematic lighting, Dramatic lighting, Golden hour sunlight, Neon lights, Studio lighting, etc.
• Painter: Numerous well-known painters including Vincent van Gogh, Claude Monet, Rembrandt, Salvador Dali, Pablo Picasso, Leonardo da Vinci, etc.
• Art Movement: 21 movements including Abstract Expressionism, Art Deco, Cubism, Impressionism, Surrealism, etc.
• Rendering engine: 20 engines including Unreal Engine, Unity, Blender, 3ds Max, Maya, Cinema 4D, Houdini, etc.
• Perspective: 20 perspectives including Aerial perspective, Bird's-eye perspective, Fish-eye perspective, Macro perspective, Worm's-eye perspective, etc.
• Material: 29 materials including Made of bronze, Made of marble, Made of glass, Made of wood, etc.
• Camera: 20 camera types including DSLR camera, Medium format camera, Drone camera, Pinhole camera, etc.
• Filter: 20 filters including Polarizing filter, Neutral density filter, Infrared filter, Vignette filter, etc.
• Scene type: 18 scenes including Fantasy scene, Horror scene, Nature scene, Science fiction scene, etc.
• Advanced Parameters:
  • Chaos: Value between 0-100. Higher values favor more varied, random, and unusual generations in exchange for less reliable compositions.
  • Seed: A positive integer (between 0 and 4294967295) to keep things steady and reproducible.
  • Negative Prompting (No): Negative prompting (e.g., --no plants tries to remove plants).
  • Image Weight (IW): Sets image prompt weight relative to text weight, default 0.25.
  • Height (H) / Width (W): Sets image dimensions, values between 256 and 2034, work better as multiples of 64.
  • Quality: Rendering quality time, default 1. Higher values cost more.
  • Aspect Ratio (AR): Generates images with the desired aspect ratio (e.g., 16:9).
  • Version: Uses algorithms 1, 2, 3, or 4.
  • Stylize: Sets how strong of a 'stylization' images have, default 2500.
  • Uplight: Uses the 'light' upscaler, closer to the original image with less detail added, ideal for faces and smooth surfaces.
  • Beta: Uses a new experimental algorithm, takes longer and only produces 2 outputs. Does not work with --hd, --stylize, and --q 5.
  • HD: Uses a different, older algorithm potentially better for larger images, best for abstract and landscape prompts.
  • Same Seed: Makes the seed affect all images of the grid in the same way.
  • Video: Saves a progress video.

The app is completely free to use, open-source on GitHub, and users can run it directly from the GitHub page and inspect the source code. Target users are AI art creators and designers using Midjourney. The core advantage is the ability to generate highly customized Midjourney prompts locally without a server, with extremely detailed and intuitive parameters, greatly lowering the barrier to writing complex prompts.