The difference between a mediocre AI-generated image and one you can actually use comes down almost entirely to the prompt. A vague request produces a vague image. A specific, structured prompt that describes the subject, style, lighting, mood, composition, and technical parameters produces something you can use for a client, a campaign, or a product listing. This guide gives you 15 copy-paste prompts organized by use case, with explanations of why each element is in the prompt.
The Anatomy of an Effective Prompt
Before the prompts, here is the structure that makes them work. A strong image generation prompt typically includes these elements:
- Subject: What is in the image and what it is doing. "A woman in her 30s" not just "a person."
- Style or medium: Photography, oil painting, 3D render, watercolor, flat vector. Without this, the model guesses.
- Lighting: Studio lighting, golden hour, overcast, rim light, softbox. This changes everything about the mood.
- Composition: Close-up, overhead flatlay, wide shot, three-quarter view. Tells the model where to put the camera.
- Mood or atmosphere: Cozy, dramatic, minimal, energetic. Translates directly into color temperature and saturation choices.
- Technical parameters: Platform-specific flags like Midjourney's
--ar 16:9 --style raw --q 2.
You do not need every element for every prompt. But the more specific you are on the elements that matter for your use case, the more predictable and reusable your results become.
Product Photography Prompts
Prompt 1 -- Minimalist product shot: A matte black ceramic coffee mug on a pale concrete surface, single sprig of eucalyptus to the right, one overhead studio light casting a clean soft shadow, commercial product photography, 4:5 ratio, no text, white negative space on left for copy --ar 4:5 --style raw --q 2
Prompt 2 -- Premium skincare flatlay: Top-down flatlay of a white glass serum bottle, three rose petals, a small white marble slab, eucalyptus leaves, pale pink background, soft natural light from the upper left, beauty product photography, luxury feel, high resolution, sharp focus throughout --ar 1:1 --style raw
Prompt 3 -- Outdoor gear lifestyle shot: A weathered brown leather hiking boot resting on a mossy granite boulder, dense Pacific Northwest forest in the soft-focus background, overcast natural light, editorial outdoor photography, authentic and rugged, no studio feel, shallow depth of field on the boot lace --ar 3:2
Portrait and People Prompts
Prompt 4 -- Professional headshot: Business headshot portrait of a 40-year-old South Asian man in a charcoal wool blazer, confident relaxed expression, clean white background, rim lighting on left shoulder, shot with 85mm lens at f/1.8, photorealistic, LinkedIn profile quality --ar 1:1
Prompt 5 -- Lifestyle editorial: A woman in her late 20s working at a minimalist standing desk, noise-cancelling headphones, natural afternoon light from a large left-side window, green potted plants on a shelf behind her, editorial lifestyle photography, warm and productive atmosphere, photorealistic --ar 4:5
Interior Design and Real Estate Prompts
Prompt 6 -- Modern Scandinavian living room: Interior design photography of a Scandinavian living room, white walls with warm oak floors, a pale linen sofa facing a low oak coffee table, a single large fiddle-leaf fig in the far left corner, afternoon light streaming through floor-to-ceiling windows, minimal decor, no clutter, architectural photography quality --ar 16:9
Prompt 7 -- Virtual staging: empty bedroom: Photo-realistic virtual staging of an unfurnished 14 x 12 bedroom with hardwood floors and two west-facing windows, styled with a king platform bed in muted sage green linen, matching light oak nightstands, a linen armchair reading nook in the far right corner, warm ambient lighting, real estate photography --ar 4:3
Marketing and Social Media Prompts
Prompt 8 -- Coffee brand banner: Wide-format lifestyle banner for a specialty coffee brand, top-down shot of a ceramic pour-over dripper mid-brew over a glass carafe, scattered whole coffee beans, a folded white linen napkin, warm oak surface, morning light from the left, terracotta and cream color palette, commercial photography, 3:1 ratio --ar 3:1
Prompt 9 -- Tech startup hero image: Modern abstract hero image for a B2B SaaS website, geometric shapes in deep navy and electric blue floating on a clean white background, subtle depth and layering suggesting connectivity and data flow, no text, no people, contemporary flat design aesthetic, digital art --ar 16:9
Prompt 10 -- Food photography: Restaurant-quality food photography of a bowl of tonkotsu ramen, rich opaque bone broth, perfectly soft-boiled egg halved to show a jammy yolk, rolled chashu pork, nori sheet, thinly sliced green onions, sesame seeds, black wooden chopsticks resting on the rim, dark moody restaurant background, shallow depth of field --ar 4:5
Illustration and Art Style Prompts
Prompt 11 -- Children's book illustration: Cute 3D render illustration of a small raccoon chef in a tiny white toque hat, stirring a bubbling copper pot in a cozy kitchen, warm amber and cream color palette, Pixar-inspired lighting and rendering style, friendly and charming, suitable for a children's book cover --ar 4:5
Prompt 12 -- Abstract art for office wall: Abstract acrylic pour painting texture, fluid swirls of deep teal, brushed gold, and ivory, high resolution, sophisticated and calm, suitable for a large-format print in a law firm or financial office, no figures, no text, landscape orientation --ar 16:9
Architecture and Exterior Prompts
Prompt 13 -- Modern home exterior: Architectural exterior render of a two-story modern farmhouse, dark charcoal board-and-batten siding, black window frames and gutters, warm interior lighting visible at dusk, native drought- tolerant landscaping with ornamental grasses, photorealistic CGI, dusk lighting with deep blue sky, magazine quality --ar 16:9
Before-and-After: Why Specificity Matters
Here is the same product shot requested two different ways. The vague version and the specific version of the same image:
VAGUE (what most people start with): "a coffee cup photo" RESULT: Generic stock-photo-looking mug, random background, flat lighting, no clear purpose or style. SPECIFIC (what you should write instead): "Overhead close-up of a white ceramic flat-white in a small cup with detailed latte art rosette, dark espresso-stained saucer, single coffee bean beside it, on a dark slate surface, moody cafe atmosphere, soft directional side lighting, shallow depth of field, commercial coffee photography --ar 1:1" RESULT: Usable product image with consistent style, clear mood, and professional lighting that matches a brand.
Negative Prompts: What to Tell the AI to Avoid
Prompt 14 -- Food flatlay with negative prompts (Stable Diffusion): Positive: "Overhead flatlay of fresh summer fruit, sliced watermelon, blueberries, yellow mango, white linen tablecloth, scattered mint leaves, bright natural light, food photography" Negative: "text, watermark, blur, distorted fruit, extra hands, unrealistic colors, oversaturated, plastic look, low quality"
Prompt 15 -- Architecture exterior with negative prompts: Positive: "Photorealistic exterior of a contemporary single-story home with floor-to-ceiling glass, warm interior glow at dusk, drought-tolerant native landscaping, clean concrete driveway" Negative: "cars, people, street signs, power lines, lens flare, overexposed sky, construction equipment, ugly neighbors"
Prompt Elements That Get Consistent Results
| Element | What to Include | Why It Matters | Example |
|---|---|---|---|
| Subject | Age, appearance, action, expression | Avoids generic placeholder figures | "woman in her 30s, smiling, mid-conversation" |
| Lighting | Source, direction, quality | Lighting changes mood more than any other element | "single softbox left, rim light right" |
| Style | Medium, era, artist reference | Sets the visual language of the image | "Pixar-style 3D render" or "editorial photography" |
| Composition | Shot type, angle, framing | Prevents random cropping and composition | "overhead flatlay" or "three-quarter front view" |
| Mood | Atmosphere, color temperature | Drives color, saturation, and overall feel | "cozy and warm" or "dramatic and high-contrast" |
| Negative prompt | What to exclude | Removes common artifacts and unwanted elements | "no text, no watermark, no blur" |
Frequently Asked Questions
Do these prompts work on Midjourney, DALL-E 3, and Stable Diffusion?
Most of the prompts work across all three platforms with minor adjustments. Midjourney responds well to the parameter flags at the end (--ar 16:9 --style raw). DALL-E 3 handles natural-language descriptions well and does not need flags. Stable Diffusion works best with shorter prompts and explicit negative prompt fields. For Stable Diffusion, break the prompt into a positive field and a negative field rather than using one long string.
Why do some of my AI-generated images still look generic?
The most common cause is a missing style or medium specification. Without telling the model whether you want photography, illustration, 3D render, or oil painting, it defaults to a middle-ground that looks like a stock photo from 2012. Adding "commercial photography" or "editorial photography" or "Pixar-style 3D render" shifts the entire visual language immediately.
How do I get consistent results across multiple images?
Consistency comes from reusing the same structural elements across all your prompts. If you are building a product line, lock in the background material, lighting setup, angle, and aspect ratio. Vary only the subject. In Midjourney, using --seed [number] can help maintain stylistic consistency across a batch. In DALL-E 3, save your base prompt as a template and only swap the product description.
What is the best AI image generator for commercial use?
For commercial use rights, DALL-E 3 (via ChatGPT Plus or OpenAI API) and Adobe Firefly offer the clearest commercial licensing. Midjourney allows commercial use on paid plans starting at $10/month, but review their terms. Stable Diffusion with the appropriate model weights (check the individual model licenses) can be used commercially for most models. Always check the license of the specific model you are using before publishing commercially.
Can I use these prompts to generate images for clients?
Yes. Save the base prompt for each client as a reusable template and adjust the subject details per shoot. Most designers and content agencies treat AI image prompts as proprietary templates, building a library of tested prompts that reliably produce on-brand results for each client. The prompt is the skill, not the button click.
Building Your Prompt Library
Save every prompt that produces a result you like. Build a personal library organized by use case: product photography, portraits, social banners, illustrations. Over time your library becomes a competitive asset. A well-tested prompt for a specific use case reliably produces professional-quality images in seconds, a capability that compounds the more prompts you collect and refine.
The fifteen prompts above are starting points. Run them, see what comes back, adjust one element at a time, and save the versions that work. That iteration process is what separates people who get mediocre AI images from people who produce images indistinguishable from professional photography or commissioned illustration.