Tutorials8 min read
Multi-Panel AI Images: Comics, Storyboards and Character Sheets
How to get four, six or nine panels in one image with the same character in all of them — prompt structure, grid sizes that hold, and how to fix drift.
On this page
- One image, many panels
- Why it's hard
- The structure that works
- Grid sizes, and which ones survive
- Four things people build with this
- When it drifts, fix it in this order
- More multi-panel work to start from
- FAQ
- How many panels can one image hold?
- How do I keep the same character across every panel?
- Can I make a comic with speech bubbles this way?
- Which model handles multi-panel best?
- What aspect ratio should I use?
One image, many panels
A four-panel comic. A nine-shot storyboard. A six-expression character sheet. A 3×3 product grid.
These all ask the model for the same difficult thing: produce several distinct frames inside a single image, and keep one subject consistent across all of them. It is a very different problem from generating one good picture, and it is where models separate most sharply — a model that nails a single portrait will happily give you nine portraits of nine different people.
This guide covers what actually works: how to structure the prompt, which grid sizes hold together, and where it falls apart.
Why it's hard
A single-image prompt asks for a look. A multi-panel prompt asks for a look plus a rule that has to hold across the whole canvas: same face, same outfit, same lighting, different pose.
Models handle rules less reliably than they handle moods. The common failure is drift — panel one and panel two match, panel six is a different person. The whole craft of multi-panel prompting is making the rule impossible to ignore.
Why this suits Nano Banana
In our gallery corpus, conditional consistency is where Nano Banana separates from the other models — the same reason it holds selective colour or a fixed character across a set. Nearly every strong multi-panel piece in our library was generated with it.
The structure that works
Four elements, in this order. Dropping any one of them is where most attempts fail.
- State the grid explicitly, with a number. "6-panel grid, 2 rows × 3 columns" beats "several panels." The model needs a count, not a vibe.
- Describe the subject once, in detail, before the panels. This is the anchor. Everything the panels share — face, hair, outfit, age, build — belongs here and nowhere else.
- Give each panel one variable, and only one. Panel 1: smiling. Panel 2: surprised. If a panel changes both expression and camera angle and outfit, drift starts there.
- Lock the invariants at the end. "Same character, same outfit, same lighting in every panel, consistent art style throughout." Redundant to a human reader; load-bearing for the model.
[6-panel grid, 2 rows × 3 columns, even borders]
[Character] A young woman, shoulder-length black hair, round
glasses, cream knit sweater, warm studio lighting
[Panels]
1. Neutral, facing camera
2. Laughing, head tilted
3. Surprised, eyebrows raised
4. Thinking, hand near chin
5. Annoyed, side glance
6. Soft smile, eyes closed
[Consistency] Identical character, outfit and lighting in every
panel. Consistent illustration style throughout. Clean white
gutters between panels.
Pixar-Style 6-Panel Expression Grid
Nano Banana ProPrompt▾
A 3D Pixar-style animated young male character showcased in a 6-panel grid layout, each panel featuring a different facial expression and outfit. Bright solid color backgrounds (yellow, blue, red, pink, purple, green). The character has soft brown wavy hair, expressive large eyes, smooth skin, and stylized cartoon proportions.
Grid sizes, and which ones survive
From what holds together in our library:
| Grid | Reliability | Best for |
|---|---|---|
| 2×2 (4 panels) | Highest | Character sheets, before/after, seasonal sets |
| 2×3 / 3×2 (6 panels) | High | Expression sheets, product angles |
| 3×3 (9 panels) | Moderate | Storyboards, sticker sheets — expect to re-roll |
| 4×4 (16 panels) | Low | Rarely holds; split into two 8-panel images instead |
The practical ceiling is around nine. Past that, each panel gets too few pixels for the model to keep detail consistent, and faces start to wander.
One Image Generates 9 Different Shots
Nano Banana ProPrompt▾
<instruction> Analyze the entire composition of the input image. Identify all key subjects present (whether a single person, group/couple, vehicle, or specific object) and their spatial relationships/interactions. Generate a coherent 3x3 grid "Contact Sheet" showcasing 9 distinct shots of exactly these subjects within the same environment. You must adapt standard cinematic shot types to fit the content (e.g., if a group, keep the group together; if an object, frame the entire object): Row 1 (Establishing Context): 1. Extreme Long Shot (ELS): Subjects appear very small within a vast environment. 2. Long Shot (LS): Full subject or group visible from top to bottom (head to toe / wheels to roof). 3. Medium Long Shot (American Shot/Three-Quarter): Framed from above the knees (for people) or a 3/4 view (for objects). Row 2 (Core Coverage): 4. Medium Shot (MS): Framed from above the waist (or the central core of an object). Focus on interaction/action. 5. Medium Close-Up (MCU): Framed from above the chest. Intimate framing of the primary subject(s). 6. Close-Up (CU): Tight framing on the face or the "front" of an object. Row 3 (Detail & Angles): 7. Extreme Close-Up (ECU): Macro detail with intense focus on a key feature (eyes, hands, logo, texture). 8. Low-Angle Shot (Worm's-Eye View): Looking up at the subject from ground level (spectacular/heroic feel). 9. High-Angle Shot (Bird's-Eye View): Looking down on the subject from above. Ensure strict consistency: The same people/objects, same clothing, and same lighting across all 9 panels. Depth of field should vary realistically (background blur in close-ups). </instruction> A professional 3x3 cinematic storyboard grid containing 9 panels. This grid showcases the specific subject(s)/scene from the input image across a comprehensive range of focal lengths. Top Row: Wide environmental shots, full view, 3/4 crop (above-knee view). Middle Row: Above-waist view, above-chest view, face/front close-up. Bottom Row: Macro detail, low angle, high angle. All frames feature photorealistic textures, consistent cinematic color grading, and correct framing for the specific number of subjects or objects analyzed.
Four things people build with this
Character sheets. One character, multiple poses or expressions, for reference or for a project bible.
Cat-Eared Student Council Character Reference
GPT ImagePrompt▾
{
"type": "anime character reference sheet",
"style": "clean polished light novel / visual novel design sheet, white background with pale lavender ornamental border, high-detail cel shading, elegant school-themed layout, soft cool lighting, crisp line art",
"character": {
"name": "霧島 澪奈",
"role": "serious female class committee president with black cat traits",
"age": "high school student, second year",
"species traits": [
"2 black cat ears on top of head with soft inner fur",
"1 long black cat tail"
],
"appearance": {
"gender presentation": "female",
"build": "slender, refined, modestly curvy",
"height note": "tall for a schoolgirl",
"hair": {
"color": "deep blue-black",
"style": "very long straight hair with slight waves at the ends, center-parted bangs, long sidelocks, half-up back section tied with a ribbon bow"
},
"eyes": {
"color": "golden amber",
"shape": "sharp but elegant"
},
"expression base": "cool, intelligent, reserved"
},
"outfit": {
"count": 9,
"pieces": [
"navy school blazer with white piping",
"white dress shirt",
"large striped ribbon tie in navy and burgundy",
"school crest on blazer chest",
"red armband with white Japanese text indicating committee leader",
"blue plaid pleated skirt",
"black semi-sheer tights",
"black loafers",
"thin gold-framed glasses"
]
},
"props": {
"count": 1,
"items": [
"dark school folder or notebook with crest"
]
}
},
"layout": {
"format": "landscape character design board",
"sections": [
{
"title": "プロフィール",
"position": "left column",
"count": 1,
"labels": ["bio text block"]
},
{
"title": "SIDE / BACK",
"position": "upper center",
"count": 3,
"labels": ["front full body", "side full body", "back full body"]
},
{
"title": "表情バリエーション",
"position": "upper right",
"count": 8,
"labels": ["通常（真面目）", "微笑み", "注意・叱責", "驚き", "照れ（目をそらす）", "困り顔", "優しい笑顔", "猫っぽく甘える"]
},
{
"title": "ディテール・アクセサリー",
"position": "middle right left panel",
"count": 5,
"labels": ["委員長の腕章", "校章", "猫モチーフのヘアピン", "リボンタイ", "細フレーム眼鏡"]
},
{
"title": "制服ディテール",
"position": "middle right right panel",
"count": 3,
"labels": ["chest and tie close-up", "waist and blazer trim close-up", "sleeve cuff and buttons close-up"]
},
{
"title": "カラーパレット",
"position": "lower left",
"count": 5,
"labels": ["髪", "瞳", "制服", "アクセント", "肌"]
},
{
"title": "特徴的なポーズ・仕草",
"position": "lower center",
"count": 4,
"labels": ["指示を出す（委員長モード）", "資料を確認する", "考え事をする", "休み時間の素顔（少し気が抜けた瞬間）"]
},
{
"title": "その他設定・ポイント",
"position": "lower right",
"count": 2,
"labels": ["classroom vignette", "moonlit night vignette"]
}
]
},
"text blocks": {
"headline": "霧島 澪奈",
"subheadline": "学級委員長",
"quote": "みんなのために、きちんとしなきゃ。それが、私の役目だから。",
"body text": "dense Japanese profile text with school year, birthday, height, blood type, club or committee, personality notes, likes and dislikes"
},
"composition notes": [
"main full-body figure at left, holding glasses near face and carrying a folder",
"side and back character turnaround in the center",
"all portraits and close-ups neatly framed in pale lavender boxes with decorative corner motifs",
"facial close-ups cropped at shoulders",
"bottom-right vignette images show the character in a classroom and outdoors at night under a crescent moon",
"overall palette is navy, black, lavender, white, burgundy, muted gold, and soft skin tones"
]
}Storyboards. A sequence rather than a set — panels read in order, so the prompt should number them as beats.
Cinematic Photo Storyboard
Nano Banana ProPrompt▾
A 3x3 cinematic storyboard contact sheet consisting of 9 distinct panels arranged in a grid. The sequence features a young woman with platinum blonde hair in a frozen alpine winter setting. The panels display various angles and shots: Close-ups: Focusing on her rosy cheeks, blue-grey eyes, and snowflakes on her eyelashes. Medium shots: Showing her wrapped in a black wool coat and blue knit scarf, holding a bouquet of dried white flowers. Wide shots: Capturing her standing alone on the frozen lake with towering snowy mountains in the background. The lighting is consistent moody blue-hour twilight across all frames. High-quality film photography aesthetic, photorealistic, 8k resolution, coherent character and color grading.
Product grids. Same object, multiple angles or colourways, laid out for a page or a listing.
Premium 3x3 Grid Studio Product Storyboard
Nano Banana ProPrompt▾
{"OverallLayout":{"Format":"3×3 grid storyboard, nine equal panels on 4:5 ratio canvas","Style":"High-end designer mockup, editorial, curated, design-driven","Focus":"Form, composition, materiality, visual rhythm over lifestyle narrative","ProductConsistency":"Exact same product, packaging, branding, materials, colors, proportions maintained across all nine panels","ProductRecognition":"Product clearly recognizable in every frame, label/logo/proportions unchanged"},"Frame1_HeroShot":{"Type":"Front-facing hero shot","Set":"Clean studio setup","Background":"Neutral, balanced composition","Product":"Nano Banana Pro package center","Lighting":"Calm, confident presentation","Camera":"Straight-on angle"},"Frame2_CloseUp":{"Type":"Close-up middle focus","Focus":"Surface texture and materials","Details":{"Texture":"Matte finish package texture","Print":"Crisp banana illustration details","Seal":"Re-sealable zipper mechanism texture","Branding":"Logo typography and colors"},"DepthOfField":"Shallow, product in sharp focus"},"Frame3_BrandEnvironment":{"Type":"Environment fit for brand","Setting":"Studio setting inspired by packaging colors","Colors":"Yellow banana tones with soft green accents","Props":{"Background":"Soft yellow backdrop with banana leaf shadows","Surface":"Matte white table with subtle texture"}},"Camera&Style":{"Quality":"Ultra high-quality studio imagery, real camera look","ColorPalette":"Soft yellow, white, green banana tones with matte finishes","Mood":"Clean, modern, approachable yet premium"},"Output":{"Format":"3×3 grid, no borders, no text, no captions, no watermarks","Quality":"High-resolution portfolio-ready"}}Editorial and sticker sets. Fashion four-ups, seasonal sticker sheets, campaign grids.
K-Beauty Four Panel Editorial
Nano Banana ProPrompt▾
Beauty editorial collage, 4-panel split-screen grid layout with thick white diagonal borders. Young woman with clean girl / K-beauty aesthetic. Flawless, dewy, 'glass skin' texture, high luminosity. Short brown bob cut with wispy bangs, smooth texture. Soft grey/hazel contacts. No-makeup makeup look with soft rosy blush high on cheeks, glossy pink lip tint, groomed natural fluffy brows. Brown houndstooth pattern newsboy cap. Small delicate gold butterfly stud earrings, thin gold chain necklace. Top left: Close-up 3/4 angle, focus on cheek highlight and earring. Top right: Direct front-facing portrait, symmetrical composition. Bottom left: Portrait with hat, 3/4 turn, looking at camera. Bottom right: Extreme close-up side profile. 8K resolution, soft commercial studio lighting, ring light reflection in eyes, desaturated background, warm skin tones, high contrast, high-end beauty photography.
When it drifts, fix it in this order
- Reduce the panel count. Nine to six solves most drift outright.
- Make the character description more specific. "A woman" drifts. "A woman, late twenties, square jaw, shoulder-length black hair with blunt fringe, silver hoop earrings" holds.
- Cut variables per panel. If each panel changes pose and angle, fix the angle for all of them.
- Restate the invariants. Adding "identical face in all panels" at the end genuinely changes the output.
- Only then, use a reference image. Feeding a portrait as reference is the strongest lock available, and worth reaching for before you give up.
Don't ask for text in the panels
Speech bubbles and captions inside a multi-panel image are the hardest combination there is — you are asking for layout consistency and text rendering at once. Generate the panels clean, add the text in any editor afterwards. If the words must be generated, that is a job for GPT Image rather than Nano Banana.
More multi-panel work to start from
✕ Bad
comic strip of a detective entering a room, finding a clue, and leaving✓ Good
three equal vertical panels, same detective and brown coat in every panel; panel 1 enters the room, panel 2 lifts a red key, panel 3 exits holding that same key; fixed camera height and lightingFAQ
How many panels can one image hold?
Four and six are reliable. Nine works but expect to re-roll. Past nine, each panel gets too little resolution for the model to hold detail, and consistency collapses — generate two images instead.
How do I keep the same character across every panel?
Describe the character once, in specific physical detail, before the panels; give each panel exactly one variable; and restate "identical character in every panel" at the end. If it still drifts, supply a reference image.
Can I make a comic with speech bubbles this way?
Generate the artwork without text and add lettering afterwards — that is what most people end up doing. Asking one model for panel consistency and legible text simultaneously is the least reliable request in this whole area.
Which model handles multi-panel best?
Nano Banana, in our library — conditional consistency across a canvas is its clearest strength. GPT Image is the better choice when the panels themselves must contain readable text.
What aspect ratio should I use?
Match it to the grid: 1:1 for 2×2 and 3×3, 4:5 or 3:4 for 2×3 portrait layouts, 3:2 for wide storyboards. A mismatched ratio squeezes panels and is a common cause of dropped detail.
LocalBanana Team
We run Nano Banana, GPT Image and other models side by side, and publish every prompt in our gallery. These guides are written from that corpus.
Updated @LocalBanana_io
