Visual Prompt Generator & Parameter Compiler
Build Production-Grade Photographic & Generative Art Prompts in Seconds
Eliminate prompt guesswork and outdated keyword spam. Select real-world camera optics, lighting geometry, historical film stocks, and composition grids to instantly compile syntactically valid prompts formatted specifically for Midjourney v6.1, FLUX.1, SDXL, or DALL-E 3.
1. Subject & Environmental Framing
2. Camera Optics & Analog Medium
3. Lighting Direction & Color Temperature
4. Composition Geometry & Engine Parameters
Midjourney structures prompts best when parameters (--ar, --v, --stylize) are strictly isolated with single spaces at the very end of the string.
The 4 Pillars of Diffusion Prompt Engineering
Modern diffusion models (FLUX.1, Midjourney v6.1) were trained on extensive photographic catalogs and cinematic datasets. They don't require generic adjectives like "hyperrealistic 8k"—they require concrete physical optical specifications.
1. Optical Hardware
Specify exact focal lengths and lenses (e.g. Sony 85mm f/1.4 for portraits, Canon 24mm Tilt-Shift for architecture) to control depth of field and barrel distortion.
2. Lighting Geometry
Name the exact lighting setup (e.g. Rembrandt 45° triangle, Clamshell beauty dish, Low golden hour rim-light) to sculpt subject shadows accurately.
3. Analog Medium
Chemical film stocks like Kodak Portra 400 and Ilford HP5 instruct the latent diffusion solver to produce organic grain and authentic tonal curves without plastic skin.
4. Engine Tokens
Append precise flags (e.g. --v 6.1 --ar 16:9 --stylize 200) isolated at the end of the prompt to prevent CLI argument parse errors.
Calibrate Output Framing with SVG Grids
After generating your image, test focal point placement against the Rule of Thirds and Fibonacci Golden Spiral.
Explore Model-Specific Parameter Silos
Deep-dive into individual model parameter tables for Midjourney, FLUX.1, Claude 3.7, ChatGPT, and SDXL.