Skip to main content
Tutorial

A Simple Prompt Formula for Stunning Images

Stop guessing at prompts. This six-part formula gives you a reliable structure that turns a vague idea into a striking, intentional image.

June 16, 2026

A Simple Prompt Formula for Stunning Images - AI image and video guide preview from eaxy (prompt formula for images)
A Simple Prompt Formula for Stunning Images - AI image and video guide preview from eaxy (prompt formula for images)

A prompt formula is a fixed structure for describing what you want, so you cover every ingredient that shapes an image instead of leaving the model to fill in the blanks. The reliable version has six parts: subject, setting, style, lighting, composition and detail. Cover those, in roughly that order, and your results jump from random to intentional.

The six-part formula

Think of it as a checklist you run through every time:

  1. Subject — the main thing and what it is doing. "A barista pouring latte art."
  2. Setting — where it happens. "in a sunlit specialty coffee shop."
  3. Style — the medium and look. "editorial food photography" or "flat vector illustration."
  4. Lighting — the single biggest lever for mood. "soft window light, warm tones."
  5. Composition — framing and angle. "close-up, shallow depth of field, rule of thirds."
  6. Detail — the finishing touches. "steam rising, ceramic cup, cozy atmosphere."

Stitched together: "A barista pouring latte art in a sunlit specialty coffee shop, editorial food photography, soft window light and warm tones, close-up with shallow depth of field, steam rising from a ceramic cup, cozy atmosphere." That single sentence gives the model everything it needs to be deliberate rather than generic.

Why order matters

Most image models weight the earlier words slightly more heavily, so lead with what matters most — the subject and its action. Push style and lighting next, because they set the entire feel, and leave fine details for last. If you bury "a red bicycle" at the end of a long prompt, the model may treat it as an afterthought.

This is also why you should not pad prompts with empty adjectives. "Beautiful, amazing, stunning, masterpiece" tells the model almost nothing. "Backlit, golden hour, 85mm, shallow depth of field" tells it a great deal. Specificity beats enthusiasm every time — a point we make in text-to-image, explained.

Lighting is your secret weapon

If you only upgrade one part of your prompts, make it lighting. The same subject reads completely differently under:

  • Golden hour — warm, soft, flattering, nostalgic.
  • Soft studio light — clean, professional, product-ready.
  • Neon / cyberpunk — moody, saturated, futuristic.
  • Candlelight — intimate, dramatic, low-key.
  • Overcast daylight — even, natural, documentary.

Naming the light is the fastest path to images that look art-directed instead of accidental.

Composition: tell it how to frame the shot

The model will frame the scene however it likes unless you guide it. Useful phrases include close-up, wide establishing shot, overhead / flat lay, low angle, eye level, rule of thirds, centered, negative space. Pair your composition choice with the right aspect ratio so the framing actually fits the canvas — see AI image aspect ratios and sizes for which shape to pick.

A worked example you can adapt

Start vague, then layer the formula:

  • v1 (vague): "a sneaker" → unpredictable.
  • v2 (subject + setting): "a white running sneaker on a concrete ledge" → better, but flat.
  • v3 (+ style + lighting): "...product photography, soft directional studio light" → looks intentional.
  • v4 (+ composition + detail): "...three-quarter angle, shallow depth of field, subtle reflection, minimal background, 4K" → ad-ready.

Each step changes one or two ingredients, which is exactly how you should iterate: adjust a single part, regenerate, compare.

Tips that make the formula even stronger

  • Use a style pack. eaxy's 30+ curated packs handle the "style and lighting" parts for you — pick portrait, product, cinematic or anime and your prompt can stay short.
  • Generate in batches. A few variations per prompt give you something to choose from and to refine against.
  • Lock what works. When a result is close, keep the structure and change only the part that is off.
  • Mind the ratio. Compose for the shape you will export in.
  • Plan for motion. If the image is destined to become video with Kling 3, a clean, well-composed still animates far better.

The short answer

Use the six-part formula — subject, setting, style, lighting, composition, detail — lead with the subject, treat lighting as your biggest lever, and change one ingredient at a time when iterating. It turns prompting from guesswork into a repeatable craft. Put it to work now: start creating and build your first prompt with the formula.

Frequently asked questions

What is a prompt formula?+

It is a fixed structure for writing prompts — a checklist of ingredients like subject, style and lighting — so you cover everything that matters instead of leaving the model to guess.

Do longer prompts always make better images?+

No. Clarity beats length. A focused prompt that names the subject, style, lighting and composition usually beats a long, rambling one full of vague adjectives.

What order should the words go in?+

Lead with the subject and action, then setting, then style and lighting, then composition and finishing details. The model weights the early words a little more heavily.

Should I use commas or full sentences?+

Both work. Short comma-separated phrases are easy to edit one piece at a time, while sentences can read more naturally. Pick what is easier for you to iterate on.

What if the image still is not right?+

Change one ingredient at a time — adjust the lighting, then the composition — rather than rewriting the whole prompt. It is the fastest way to learn what each part does.

Make it with eaxy

Describe anything and generate stunning images in seconds — then bring them to motion with Kling 3.

Related

Useful next steps