Skip to main content
Nano Banana AI

Nano Banana

How to Write Nano Banana Prompts? 5 Frameworks to Make You a Pro

Nano Banana AI4 min read

Nano BananaPromptsText to ImageFrameworks
Nano Banana prompt frameworks cover: structured writing recipes with a polished image preview
Contents+

Good Nano Banana prompts aren’t longer—they’re clearer. Instead of stacking adjectives, use a reusable scaffold so the first frame lands on the right subject, light, and mood. Then refine with one focused chat change at a time. These five frameworks cover everyday creator jobs: portraits, products, atmospheric scenes, series consistency, and reference-driven edits.

Practice as you read on nanobanana-ai.us.

Workspace example: https://app.nanobanana-ai.us/en/generate/image-tools/nano/

Framework 1: Subject-first (your default)

Most weak Nano Banana text-to-image results fail for a simple reason: the model never locks onto who or what before style words flood in. Subject-first keeps this order:

  1. Subject — person, product, or object (visible traits)
  2. Action / state (optional)
  3. Scene
  4. Style
  5. Light + lens

Example

A young woman in a linen shirt by a café window, soft daylight from the left, shallow depth of field, 35mm, realistic portrait photography

Rule of thumb: observable facts first, vibe words second. That beats “beautiful, masterpiece, 8k” for Nano Banana.

Framework 2: Scene story (mood boards & narrative frames)

When you want a picture that feels like a beat in a story, lead with narrative:

  1. Time / weather
  2. Place
  3. What the subject is doing
  4. A few environment details (two or three is enough)
  5. Camera language (shot size, DOF, motion feel)

Example

Dusk after rain in a city alley, a short-haired woman walking with a clear umbrella, neon reflected on wet asphalt, distant silhouettes, cinematic contrast, 85mm shallow depth of field

Don’t dump ten background props. The goal is readable subject + locked atmosphere.

Framework 3: Light & lens (photo-real portraits & products)

For a “shot on camera” look, treat light and optics as hard constraints:

  1. Key light direction (side, backlight, top, window)
  2. Softness / color temperature
  3. Focal length / depth of field
  4. Framing (eye-level, top-down, close-up)
  5. Subject–background relationship (bokeh, seamless, table material)

Portrait

Window side-light portrait, soft shadows, natural skin texture, 85mm, shallow depth of field, warm bokeh background

Product

A ceramic coffee cup on a walnut table, high window side light, soft shadows, realistic product photography, shallow depth of field, clean background

This framework shines for e-commerce heroes, covers, and premium still life.

Framework 4: Style anchor (series that stay consistent)

Series posters, character sheets, and brand kits fall apart when every prompt reinvents the look. Fix a style anchor line, then swap only subject and scene:

  1. Lock the style (photoreal / flat illustration / cinematic / cyber)
  2. Lock light preferences
  3. Change only subject, outfit, or place each time
  4. Optionally add “same style as the previous image”

Reusable anchor

Flat illustration, clean linework, limited palette, soft shading, centered composition

Apply it like:

[Subject and scene]. Flat illustration, clean linework, limited palette, soft shading, centered composition

For Nano Banana series work, anchors beat rewriting full prompts from scratch.

Framework 5: Keep / replace (references & chat edits)

With a reference image—or when refining a first frame—don’t re-describe the whole world. Say two things:

  1. What to keep (face, hair, product shape, composition)
  2. What to change (background, outfit color, lighting, props)

Reference example

Keep the same facial features and hairstyle; replace only the background with a dusk beach; leave outfit and pose unchanged

Chat refine example

Keep the subject; soften the key light slightly and lower contrast a bit; leave everything else

That’s where Nano Banana chat editing wins: one variable per turn, controllable results.

Which framework when

Goal Start with
Fast usable first frame Subject-first
Story / atmosphere Scene story
Photographic polish Light & lens
Consistent series look Style anchor
Reference or revision Keep / replace

Mix freely—for example subject-first + light & lens for the first frame, then keep / replace in chat.

Common pitfalls

  1. More adjectives than facts — lead with materials, color, and pose.
  2. Too many changes in one turn — outfit + background + light together cause drift.
  3. Conflicting style cues — “minimal” and “ornate baroque” fight each other.
  4. No anchor for series — color and line quality wander frame to frame.

Wrap-up

How to write Nano Banana prompts isn’t about length—it’s about structure. Keep these five frameworks in your toolkit: subject-first, scene story, light & lens, style anchors, and keep / replace. Pick the right scaffold for the first frame, then refine one change at a time.

Open nanobanana-ai.us, adapt any example above, and generate one frame yourself—muscle memory beats theory.

Ready when you are

Start generating with Nano Banana

Open the workbench and use Gemini 2.5 Flash Image for text-to-image and conversational editing. Prompts, capability guides, and tutorials live on nanobanana-ai.us—accelerate creation from here.

Related articles