Your Midjourney Adventure: Adding Oomph To Your Prompts with ChatGPT

Introduction

Discover Midjourney and ChatGPT, two powerful tools that can revolutionize your creative process for generating images.

Midjourney is an AI-powered image generation platform, brings your wildest imaginations to life, while ChatGPT, a state-of-the-art language model, crafts compelling prompts to guide Midjourney’s visual creation.

As you begin this exciting journey, you realize the endless possibilities that lie ahead. By leveraging ChatGPT’s linguistic skills and Midjourney’s visual generation capabilities, you can push the boundaries of AI-assisted art and storytelling.

You will learn the art of prompt engineering, mastering the interplay between language and visuals to create stunning and emotionally evocative images. All the variations possible with MidJourney are discussed in this post, along with the AI recipes.

With Midjourney and ChatGPT as your trusty companions, you are ready to explore uncharted territories and become a pioneer in the fascinating field of AI-powered creativity.

The world awaits your imagination, and the journey promises to be an unforgettable adventure.

The Importance of Prompts in Midjourney

In Midjourney, the more specific and descriptive your prompts are, the better the AI can understand and visualize what you’re aiming for.

Each keyword acts as a brush stroke, adding depth, color, and texture to the final masterpiece. By carefully selecting words that encapsulate the essence of your vision, you provide the AI with a clear roadmap to follow.

For instance, instead of simply saying “a castle,” try “a majestic, medieval stone castle with towering spires and a drawbridge, surrounded by a misty moat.” These details paint a vivid picture, guiding the AI to create an image that aligns closely with your imagination.

Guiding the AI to create desired visuals
Prompts are the language you use to communicate with the AI, and the more fluent you become, the better you can convey your ideas. By mastering the art of prompt crafting, you gain the power to steer the AI towards your desired outcomes.

Think of it as a collaborative dance between you and the AI. Your prompts provide the steps, and the AI follows your lead, interpreting and bringing your words to life in stunning visual form. The clearer and more precise your guidance, the more likely the AI is to generate images that match your expectations.

Leveraging ChatGPT as a Prompt Genius

By explicitly instructing ChatGPT to take on the role of a generative image genius, you prime it to think and respond in ways that help with creating compelling image prompts.

This sets the stage for a productive collaboration, with ChatGPT ready to apply its vast knowledge to your image generation.

You can ask ChatGPT to suggest unique combinations, to elaborate on specific themes, or to provide examples of prompts in different genres.

The more you interact with ChatGPT, the more it learns about your preferences and style, allowing it to tailor its suggestions to your needs.

A Prompt Template for MidJourney

Here is a structured prompt template that you can use with ChatGPT to generate prompts for MidJourney:

I want you to act as a master prompt creator for Midjourney. Your task is to help me craft highly detailed and imaginative prompts that will yield stunning AI-generated images.

[Subject] with [Key Attributes], in [Style and Mood], featuring [Colors and Lighting], including [Optional Elements], captured with a [Camera Make], using [Camera angles, settings]. [Exclusions].

This template provides a framework for creating comprehensive, well-structured prompts that cover all the essential aspects of an image.

Important: Once you write the full image prompt, get ChatGPT to make the prompt cryptic, without losing its meaning, to under 400 characters.

Let’s break it down.

[Subject]: This is the main focus of your image, such as a person, object, or scene. Be as specific as possible.

[Key Attributes]: Describe the distinctive qualities or characteristics of your subject. This could include textures, patterns, or unique features.

[Style and Mood]: Specify the artistic style (e.g., realistic, abstract, cartoonish) and the emotional atmosphere you want to convey (e.g., whimsical, mysterious, serene).

[Colors and Lighting]: Define the color palette and lighting conditions that will set the tone for your image. Consider the time of day, light sources, and color schemes.

[Optional Elements]: Add any additional objects, creatures, or environmental details that will enhance your image and make it more compelling.

[Camera Make] and [Camera angles, settings]: Specify the type of camera and the photographic techniques you want to emulate, such as wide-angle, macro, or long exposure shots.

[Exclusions]: Clearly state any elements you don’t want to appear in your image to avoid unwanted results.

By filling in each component of this template, you create a robust prompt that provides the AI with a comprehensive understanding of your vision, increasing the likelihood of generating an image that exceeds your expectations.

To help you get started on the right note, I have included a set of keys for each variable in the prompt above.

Optional Elements in MidJourney

MidJourney offers a range of settings and parameters that can significantly influence the final output. These include options like aspect ratio, quality, stylize, and chaos, each of which can be adjusted to suit your preferences.

Each of these is detailed below with an example.

Aspect Ratio (–ar or –aspect)
The aspect ratio is the width-to-height ratio of an image, expressed as two numbers separated by a colon (e.g., 16:9, 4:3, 1:1). It determines the proportions of the generated image.

Aspect Ratio Example
/imagine prompt: A stunning landscape –ar 16:9
This will create a wide, cinematic image with a 16:9 aspect ratio.

Quality (–q or –quality)
The quality parameter determines the amount of time spent generating an image, affecting its level of detail. Higher quality settings (up to 1) produce more detailed results but take longer to process. Lower settings (down to 0.25) are faster but may have less detail.

Quality Example
/imagine prompt: A detailed architectural rendering –q 1
This will generate a high-quality, detailed image of an architectural design.

Stylize (–s or –stylize)
The stylize parameter controls the strength of Midjourney’s default artistic style. Higher values (up to 1000) produce more stylized and artistic images, while lower values (down to 0) result in images that more closely match the prompt.

Stylize Example
/imagine prompt: A portrait of a woman –s 500
This will create a highly stylized, artistic portrait with enhanced colors and composition.

Chaos (–c or –chaos)
The chaos parameter determines the level of randomness and variation in the generated image grid. Higher values (up to 100) produce more varied and unexpected results, while lower values (down to 0) lead to more consistent and predictable images.

Chaos Example
/imagine prompt: A surreal dreamscape –c 75
This will generate a series of images with high variation, each depicting a unique and unpredictable surreal scene.

By adjusting these parameters, users can fine-tune their Midjourney prompts to achieve the desired visual style, composition, and level of detail in their generated images.

How To Use The Variables in MidJourney

For each variable, I have given 10 examples, or Keys below which you can try out.

[Subject] =

1. A futuristic cityscape
2. A mythical creature
3. An enchanted forest
4. A surreal dreamscape
5. A post-apocalyptic wasteland
6. An underwater kingdom
7. A steampunk airship
8. An alien planet
9. A haunted mansion
10. A cyberpunk alleyway

[Key Attributes] =

1. With neon-lit skyscrapers and flying vehicles
2. With shimmering scales and glowing eyes
3. With bioluminescent flora and fauna
4. With gravity-defying structures and impossible geometry
5. With rusted, decaying ruins and makeshift shelters
6. With vibrant coral reefs and mysterious sea creatures
7. With intricate clockwork mechanisms and brass fittings
8. With otherworldly landscapes and strange lifeforms
9. With eerie, flickering candlelight and creeping shadows
10. With rain-slicked streets and holographic advertisements

[Style and Mood] =

1. In a sleek, minimalist aesthetic
2. In a whimsical, fairy-tale inspired style
3. In a dark, foreboding atmosphere
4. In a psychedelic, mind-bending ambiance
5. In a gritty, dystopian setting
6. In a serene, tranquil underwater environment
7. In a nostalgic, Victorian-era inspired design
8. In a vibrant, otherworldly color palette
9. In a gothic, horror-inspired tone
10. In a neon-noir, cyberpunk vibe

[Colors and Lighting] =

1. Featuring cool, blue and purple hues with glowing accents
2. Featuring warm, golden tones and soft, diffused light
3. Featuring deep, rich greens and mysterious, dappled sunlight
4. Featuring vivid, saturated colors and surreal, dream-like lighting
5. Featuring desaturated, earthy tones and harsh, contrasting shadows
6. Featuring bright, tropical colors and shimmering, underwater light
7. Featuring sepia tones and warm, gas-lamp lighting
8. Featuring an iridescent, alien color scheme and strange, bioluminescent light
9. Featuring muted, ghostly hues and eerie, moonlit shadows
10. Featuring neon pinks and blues with high-contrast, dramatic lighting

[Optional Elements] =

1. Including advanced technology and holograms
2. Including ancient ruins and mystical symbols
3. Including fairy-tale creatures and enchanted objects
4. Including Escher-like staircases and portals
5. Including abandoned vehicles and overgrown vegetation
6. Including sunken treasures and lost civilizations
7. Including vintage gadgets and steam-powered machines
8. Including strange, alien artifacts and structures
9. Including ghostly apparitions and supernatural phenomena
10. Including futuristic weapons and cybernetic enhancements

[Camera Make] =

1. Hasselblad X1D-50c
2. Leica M10-R
3. Sony A7R IV
4. Nikon Z7 II
5. Canon EOS R5
6. Fujifilm GFX 100S
7. Olympus OM-D E-M1 Mark III
8. Panasonic Lumix S1R
9. Sigma fp L
10. Phase One XT

[Camera angles, settings] =

1. Wide-angle, low-key lighting, shallow depth of field
2. Telephoto, soft focus, dreamy atmosphere
3. Fisheye, high-contrast, distorted perspective
4. Macro, vibrant colors, sharp details
5. Tilt-shift, miniature effect, selective focus
6. Long exposure, smooth water, ethereal mood
7. Aerial view, golden hour, dramatic shadows
8. Dutch angle, motion blur, chaotic energy
9. Low-angle, silhouettes, mysterious ambiance
10. High-angle, symmetrical composition, minimalist aesthetic

[Exclusions] =

1. No humans or animals
2. No modern technology or devices
3. No text or logos
4. No weapons or violence
5. No urban or industrial settings
6. No cliché or overused imagery
7. No cartoonish or exaggerated features
8. No black and white or monochromatic elements
9. No references to specific brands or products
10. No low-quality or blurry components

10 Midjourney Prompts Crafted With ChatGPT Using Variables

These prompts have been crafted using the template and the variables discussed above. You can use these prompts in MidJourney, or modify these as you please before use.

1. Neon metropolis, azure titans pierce clouds. Antigrav vessels glide. Sony A7R IV captures futuristic symphony, wide-angle lens drinking cool hues. Shallow focus, low-key illumination. Devoid of organic life. –ar 16:9 –q 1 –s 500 –c 75
2. Iridescent chimera, eyes aflame. Whimsical realm bathed in golden whispers. Leica M10-R immortalizes mythic menagerie, telephoto lens weaving dreamy tapestry. Enchanted relics whisper ancient tales. Modern world banished. –ar 4:3 –q 0.5 –s 750 –c 50
3. Luminous forest, secrets pulsing. Emerald depths hide primordial ruins. Canon EOS R5 unveils bioluminescent wonders, macro lens etching vibrant details. Dappled light paints arcane glyphs. Silence reigns. –ar 1:1 –q 1 –s 250 –c 25
4. Reality warps, geometry rebels. Psychedelic maelstrom of saturated hues. Phase One XT captures Escher’s fever dream, tilt-shift lens bending perception. Staircases to nowhere, portals to everywhere. Peace prevails. –ar 2:3 –q 0.75 –s 600 –c 100
5. Civilization’s corpse, nature reclaims. Rust-kissed relics, makeshift sanctuaries. Nikon Z7 II immortalizes gritty rebirth, long exposure smoothing time’s flow. Verdant tendrils embrace decay. Urban ghosts absent. –ar 16:10 –q 0.5 –s 800 –c 60
6. Submerged utopia, coral castles rise. Enigmatic denizens glide through prismatic waters. Olympus OM-D E-M1 Mark III captures aquatic dreamscape, aerial view painting liquid gold. Lost wonders whisper below. –ar 3:2 –q 0.75 –s 500 –c 70
7. Brass leviathan sails timeless skies. Gears tick, pistons hiss. Fujifilm GFX 100S freezes steampunk odyssey, Dutch angle blurring boundaries. Sepia memories dance in gaslight glow. Reality bends, never breaks. –ar 5:4 –q 1 –s 300 –c 85
8. Alien eden, iridescent vistas unfold. Bizarre flora pulsates, structures defy comprehension. Panasonic Lumix S1R frames extraterrestrial tableau, symmetry framing cosmic minimalism. Chromatic explosion, monochrome banished. –ar 1:1 –q 0.5 –s 700 –c 95
9. Shadows writhe, spectral whispers echo. Hasselblad X1D-50c pierces veil between worlds, low angles carving silhouettes in moonlight. Phantoms dance, secrets lurk. Nameless dread lingers. –ar 16:9 –q 1 –s 800 –c 40
10. Neon rain paints cybernetic canvas. Chrome gleams, holograms seduce. Sigma fp L warps reality, fisheye lens distorting neon-noir tableau. Future bleeds into present. Clarity cuts through chaos. –ar 2:3 –q 0.75 –s 600 –c 90

Refining the Generated Images in MidJourney

Crafting the perfect image rarely happens on the first try. It’s an iterative process that involves fine-tuning your prompts based on the AI’s initial outputs.

By carefully analyzing the generated images, you can identify areas that need improvement or elements that don’t quite align with your vision.

Perhaps the colors are too muted, or the composition feels off-balance. Maybe the style isn’t quite what you had in mind, or certain details are missing. This is where the power of iteration comes into play. By making small adjustments to your prompts, you can gradually guide the AI towards the desired outcome.

Think of it as a conversation with the AI, where you provide feedback and suggestions, and the AI responds with updated visuals.

Each iteration brings you closer to your ideal image, as you learn to communicate your ideas more effectively through your prompts.

Experimenting with different combinations of these settings can lead to surprising and delightful results. For example, increasing the stylize value might add a more artistic flair to your image, while adjusting the aspect ratio can change the composition and focus.

Don’t be afraid to play around with these settings and see how they interact with your prompts. You might discover unexpected combinations that take your visuals to new heights, pushing the boundaries of what you thought was possible.

As you iterate and experiment, you’ll gradually develop a keen sense of what works and what doesn’t. You’ll start to recognize the prompts and settings that consistently produce visuals that resonate with your artistic vision.

This process of refinement is not just about technical proficiency; it’s also about developing your unique style and aesthetic. By continually honing your prompts and settings, you’ll create a body of work that is distinctly yours, reflecting your creative voice and vision.

Remember, the goal is not to create a perfect image on the first try, but rather to start a journey of exploration and discovery. Understand the process, learn from each iteration, and allow yourself to be surprised and inspired by the AI’s creations.

Conclusion

As you reflect on your journey with Midjourney and ChatGPT, you realize that it has been an exhilarating adventure filled with endless possibilities.

You’ve discovered the power of AI to enhance and amplify your creativity, pushing you to think beyond the limits of your imagination.

Through the process of crafting prompts, iterating on visuals, and experimenting with settings, you’ve developed a deep appreciation for the collaborative nature of AI-assisted art.

You’ve learned to work in tandem with these powerful tools, leveraging their strengths to bring your ideas to life in ways you never thought possible.

Next Steps

1. To help you find your way around Phewture, I have put together a set of AI Recipes under Wayfinding. Do go through these and you’ll navigate like a pro through this stream of consciousness. 😄
2. The Learning Methods are exercises that I’d recommend if you wish to wrap your head around the possibilities of using AI Recipes at work, or for play.
3. Don’t forget to leave your comments below and share your joy with your friends on social media. Use the share icons below this post to gain some good karma.

Wish to train your in-house team on AI techniques?

Phewture offers AI-spurred training for teams. Do check out the Training Services.