Character Expression Grid, made with Nano Banana 2.1
Example made with Nano Banana 2.1

Text to imageComics & stickers

Nano Banana 2.1 grid image promptCharacter Expression Grid

The exact text-to-image prompt behind this picture, made with Nano Banana 2.1.

Fill in the blanks

Change any box and the prompt below updates. Copy it, or make one with your version.

Prompt

A 3x3 expression grid of the same original character, a young woman with voluminous dark wavy hair and round wire-rimmed glasses, in a polished 3D animated-film style. The same framing in every panel: peeking through a torn hole in white paper. Each panel has a different expression and action: winking, smirking, thinking, laughing, shocked, sleepy, proud, shy and determined. Identical face, hair and proportions in all nine panels; only the expression, action and outfit colour change. Thin white gutters, no text.

Prompts stay in English, the language these models follow best. Copy it as it is, or change a few words.

Opens Nano Banana 2.1 with this prompt and the settings below filled in.

91 tested Nano Banana prompts →

Settings

Mode
Text to image
Aspect ratio
1:1
Resolution
1k
Price
From 48 credits

The exact price shows before you generate.

Make your own version

  1. Identical face, hair and proportions in all nine panels is the line that keeps one person looking the same everywhere. Describe them once, in detail, at the start.
  2. This is a fill-in template: swap the parts in braces, like {frame}, in the boxes above and keep the rest word for word.
  3. Asking for A 3x3 expression grid of the same original character gets several panels in one picture. Describe each panel in order.

Where this template comes from

Adapted from a prompt by @austinit, collected in freestylefly/awesome-gpt-image-2 under MIT. PopNow rewrote it as a fill-in template and made the example above.

Style prompt guide

Style traits to keep

A nine-panel character study renders one consistent adult woman across a neat 3x3 grid using 3D animation aesthetics. Nano Banana 2.1 excels here because it follows strict multi-panel constraints, preserving facial geometry while swapping out facial emotions cleanly inside torn-paper frames.

The modern digital feature style creates a lively, tactile look through smooth forms, clean lighting, and controlled paper framing.

Texture
Smooth, poreless skin and clean hair volumes contrast sharply with the fibrous, physical rip of the white paper borders.
Light
Soft studio illumination creates gentle rim lighting along the hair edges and adds glossy specular reflections inside the eyes.
Composition
Nine equal square apertures arranged symmetrically keep the viewer focused entirely on micro-acting and head tilts.
Palette
Vibrant, shifting shirt hues pop against the neutral torn white paper background and rich dark hair.

What defines this look

  1. layout definitionA 3x3 expression grid of the same original character

    This establishes the multi-panel matrix structure and commands the model to replicate one identity rather than generating nine random people.

  2. character and aesthetic basea young woman with voluminous dark wavy hair and round wire-rimmed glasses, in a polished 3D animated-film style.

    It locks in the recognizable features and specifies modern digital feature film rendering, complete with smooth shading and stylized forms.

  3. consistent framing deviceThe same framing in every panel: peeking through a torn hole in white paper.

    This provides a uniform framing element across the grid, anchoring the composition so the emotional shifts stand out cleanly.

  4. expression listEach panel has a different expression and action: winking, smirking, thinking, laughing, shocked, sleepy, proud, shy and determined.

    Providing an exact nine-item emotional breakdown gives the generator direct assignments for every individual square in the grid.

  5. negative constraint enforcementIdentical face, hair and proportions in all nine panels; only the expression, action and outfit colour change. Thin white gutters, no text.

    This prevents visual drift between boxes, stops unwanted typography, and ensures clean borders separate each cell.

Same style, new subjects

Corgi expression sheet

Swaps the human subject for an expressive pet character while keeping the animation shading and torn-paper setup.

Prompt

A 3x3 expression grid of the same original pet, a fluffy pembroke welsh corgi with oversized ears, in a polished 3D animated-film style. The same framing in every panel: peeking through a torn hole in white paper. Each panel has a different expression and action: happy, panting, curious, guilty, barking, sleepy, proud, startled and focused. Identical fur patterns and proportions in all nine panels; only the expression and bandana colour change. Thin white gutters, no text.

Architectural time lapse

Applies the grid and paper cutout aesthetic to an environment study showing diverse lighting scenarios.

Prompt

A 3x3 grid of the same modern street corner cafe with outdoor tables and striped awnings, in a polished 3D animated-film style. The same framing in every panel: peeking through a torn hole in white paper. Each panel has a different weather condition and time of day: sunrise, rain, midday sun, heavy snow, golden hour, fog, twilight, thunderstorm and starry night. Identical building geometry in all nine panels; only the lighting and atmosphere change. Thin white gutters, no text.

Magical potion bottles

Uses the 3D sheet presentation to showcase prop design variants and dynamic elemental materials.

Prompt

A 3x3 grid of the same round glass potion bottle with an ornate cork stopper, in a polished 3D animated-film style. The same framing in every panel: peeking through a torn hole in white paper. Each panel has a different liquid effect: swirling blue smoke, glowing green slime, sparkling gold liquid, frozen purple crystals, bubbling red lava, starry black void, electric yellow sparks, pink mist and blooming flower petals inside. Identical bottle shape in all nine panels. Thin white gutters, no text.

Model and settings

Stick to the default 1:1 aspect ratio because nine square panels naturally balance inside a square canvas. The 1k resolution handles initial generation quickly while keeping panel boundaries tidy. If you plan to crop individual panels for standalone stickers or avatar icons, generate in a higher resolution setting so the smaller crops maintain sharp edges and clear eye details.

  • Nano Banana ProSwitch to Nano Banana Pro when you want to enforce exact geometric grid placement and ensure complex panel layouts follow strict structural plans.

When the style drifts

Character features shift, making some panels look like a completely different person.
Strengthen consistency by adding specific static descriptors like exact eye color and hair length, ensuring identical facial geometry across panels.
The torn-paper cutout looks flat or turns into a plain circular frame.
Add jagged rough torn edges with physical paper grain and cast shadows behind the character to restore tactile depth.
Random gibberish text or speech bubbles appear inside the grid frames.
Make sure no text, no labels, no speech bubbles is reinforced at the end of the prompt.

Questions about this style prompt

Why did the model generate fewer than nine panels?

Complex layout prompts can occasionally merge cells. Reiterate explicit 3x3 grid with exactly nine distinct square panels separated by white borders to force the correct matrix structure.

Can I turn one of these panels into an individual avatar?

Yes. Crop any individual box out of the generated 1:1 grid using any basic photo crop tool to use as a standalone sticker or reaction image.

How do I make the 3D look more realistic instead of animated?

Replace polished 3D animated-film style with cinematic photoreal portraiture and specify natural studio lighting, realistic skin texture, and depth of field.