The Idea

Why this one works

The magic is the tension: a face caricatured hard — trapezoid head, impossibly long neck, tube-like hair — that your friends still instantly recognize as you. The carousel format multiplies it: one upload, five wildly different personalities, so people swipe to find 'their' one. It reads as premium because every style locks the same handcrafted material — matte clay skin, tactile felt hair, soft studio light — while only the shapes and background colour change.

The whole trick

Caricature the shapes as far as you dare, but lock identity: keep the real hair colour, eye colour and facial identity 'recognizable despite strong caricature'. That one line is what makes it read as YOU and not a random cartoon.

Workflow at a glance

Four steps, start to scroll-stopper.

Pick one clear photo

A single front-facing, well-lit portrait of the subject — same photo feeds all five prompts.

Lock the recipe

Keep identity + the handcrafted material (clay skin, felt hair, studio light) constant across all five.

Run the five styles

Copy each prompt below, run it on the same photo, one portrait per image.

Post as a carousel

Five 4:5 images in order — let people swipe for their favourite.

Step One

Pick the photo + subject

Everything runs off one uploaded photo, so choose it well: a clear, front-facing, shoulders-up portrait with the face and hair fully visible. The clearer the source, the better the caricature holds the likeness.

The subject

The one photo, five ways

One person from a single uploaded portrait. Across all five images, the model keeps her recognizable black hair, blue-green eyes and facial identity — that consistency is what makes the set read as the same person in five moods, not five different cartoons. Only the head/feature shapes and background colour change per style.

Recognizable, caricatured

Match the description to the photo

The prompts say 'black hair, blue-green eyes' because that's the source photo. Swap those to YOUR real hair and eye colour before running — otherwise the model fights your photo and the likeness slips.

Step Two

Lock the shared recipe

Before the five styles diverge, they all share the same handcrafted look. Lock this once and it carries the set — it's why five different shapes still feel like one premium series.

Frame	What we see	Prompt focus
THE SLEEPY ONE	Rounded trapezoid head, impossibly long neck, tiny half-closed eyes, enormous fluffy cloud-like dark hair, taupe knit sweater on a dusty-pink background.	The calmest of the set — soft, downturned, dreamy. Big cloud hair does the work.
THE SWEET ONE	Rounded head, huge glossy curious eyes, tiny triangular nose, warm smile with rose lipstick, long wavy velvet hair, black shirt on pink.	The most 'likeable' portrait — big eyes + warm smile. Usually the one people pick as theirs.
THE WILD ONE	Extremely elongated cylindrical head, long tubular nose, and an explosion of thick hollow tube-like hair strands in every direction, dark red top on sunny yellow.	The showstopper — maximum exaggeration. The tube-hair explosion is the shareable hook.

This recipe is baked into all five prompts below — keep it constant and only change the head/feature shapes and the background colour per style:

The shared recipe

[FORMAT] A NEW full-bleed 4:5 portrait image for an Instagram carousel. Centered, frontal, shoulders-up. Head and hair fully in frame; subject fills almost all the canvas. Deliver ONE portrait.

[IDENTITY] Transform the person in the uploaded photo into a high-end handcrafted 3D cartoon, retaining their recognizable hair colour, eye colour and facial identity despite strong, playful caricature. (Set the hair/eye colour to match the real photo.)

[MATERIAL & LIGHT] Matte clay skin with microtexture, tactile felt hair, soft cinematic diffused studio lighting with a delicate rim light, premium animated-feature quality. A single solid background colour.

[NEGATIVE] NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people.

Step Three

The five styles, in carousel order

Post them in this order — it opens calm and builds to the wildest, which keeps people swiping. Each slide is the same face, a different personality.

Time	Beat
Slide 1	The Sleepy One — trapezoid head, long neck, tiny sleepy eyes, giant cloud hair. Dusty-pink background.
Slide 2	The Goofy One — asymmetrical head, huge ears, misaligned sideways eyes, goofy uneven-teeth smile. Vivid red background.
Slide 3	The Sweet One — rounded head, huge glossy eyes, warm smile, rose lipstick, wavy velvet hair. Pink background.
Slide 4	The Daydreamer — perfectly spherical head, ultra-thin neck, wide capsule eyes looking sideways, mischievous closed smile. Gray background.
Slide 5	The Wild One — elongated head, long tubular nose, explosion of tube-like hair strands. Sunny-yellow background.

The prompts

Run each of these on the SAME uploaded photo — one image per prompt. They're the exact prompts that made the carousel; just swap 'black hair, blue-green eyes' for your own colours.

⬇ Download prompt guide (PDF)

Cartoon Portraits — 5 styles

[HOW TO USE] Upload one clear, front-facing, shoulders-up photo. Run each numbered prompt below on that same photo to get the matching portrait. 4:5, one portrait per image. Change the hair/eye colour to match your real photo.

[01 · THE SLEEPY ONE] Create a NEW full bleed portrait image, 4:5 aspect ratio, for an Instagram carousel. Transform the woman in the uploaded portrait into a high-end whimsical handcrafted 3D cartoon, retaining recognizable black hair, blue-green eye colour where visible, and facial identity despite strong caricature. An oversized rounded trapezoid head, impossibly long slender neck, long softly rounded rectangular nose, tiny half-closed sleepy eyes and a small downturned mouth. Enormous fluffy cloud-like dark hair with dense velvet fibers. Taupe knit sweater. Solid dusty pink background. Soft diffused studio lighting, subtle rim light, matte clay skin, tactile felt hair, premium animated feature render. Frontal shoulders-up composition, head and hair completely in frame, subject fills almost all canvas, with small blank top margin. NO text, no watermark, no inset portrait, no arrows, no graphic design. Deliver one portrait.

[02 · THE GOOFY ONE] Create a NEW full bleed 4:5 portrait image for an Instagram carousel. Transform the woman in the uploaded portrait into a high-end handcrafted 3D cartoon, retaining her recognizable black hair, blue-green eyes, and facial identity despite extreme playful caricature. An oversized asymmetrical head, thin elongated neck, huge rounded ears, long irregular nose, misaligned eyes looking sideways, raised brows and a small goofy smile with uneven oversized teeth. Long dark hair. Taupe sweater. Solid vivid red background. Soft cinematic diffused studio lighting, delicate rim light, matte clay skin with microtexture, tactile felt hair, premium animated feature quality. Centered frontal shoulders-up portrait, head and hair fully visible. NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people. Deliver ONE portrait.

[03 · THE SWEET ONE] Create a NEW full bleed 4:5 portrait image for an Instagram carousel. Transform the woman in the uploaded portrait into a high-end handcrafted 3D cartoon, retaining her recognizable black hair, blue-green eyes, and facial identity despite extreme playful caricature. An oversized rounded head, short neck, large rounded ears, huge glossy curious eyes, tiny triangular nose, soft warm smile and rose lipstick. Long dark sculpted wavy hair, plush velvet texture. Black shirt. Solid pink background. Soft cinematic diffused studio lighting, delicate rim light, matte clay skin with microtexture, tactile felt hair, premium animated feature quality. Centered frontal shoulders-up portrait, head and hair fully visible. NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people. Deliver ONE portrait.

[04 · THE DAYDREAMER] Create a NEW full bleed 4:5 portrait image for an Instagram carousel. Transform the woman in the uploaded portrait into a high-end handcrafted 3D cartoon, retaining her recognizable black hair, blue-green eyes, and facial identity despite extreme playful caricature. An enormous perfectly spherical head, impossibly long ultra-thin cylindrical neck, huge circular ears, extra-wide capsule-shaped eyes with oversized pupils looking sideways, tiny triangular nose and a mischievous asymmetric closed smile. Dark velvety sculpted hair, taupe sweater. Solid gray background. Soft cinematic diffused studio lighting, delicate rim light, matte clay skin with microtexture, tactile felt hair, premium animated feature quality. Centered frontal shoulders-up portrait, head and hair fully visible. NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people. Deliver ONE portrait.

[05 · THE WILD ONE] Create a NEW full bleed 4:5 portrait image for an Instagram carousel. Transform the woman in the uploaded portrait into a high-end handcrafted 3D cartoon, retaining her recognizable black hair, blue-green eyes, and facial identity despite extreme playful caricature. An extremely elongated cylindrical head, oversized long tubular nose with rounded tip, tiny sleepy half-closed eyes, very long slender neck and an explosion of thick hollow tube-like dark hair strands pointing in every direction. Dark red sleeveless top. Solid sunny yellow background. Soft cinematic diffused studio lighting, delicate rim light, matte clay skin with microtexture, tactile felt hair, premium animated feature quality. Centered frontal shoulders-up portrait, head and hair fully visible. NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people. Deliver ONE portrait.

[NEGATIVE (in every prompt)] NO text, no watermark, no inset portrait, no arrows, no graphic design, no extra people. One portrait per image.

Get the full method + the growing vault

✓The full course — every step, start to scroll-stopper
✓The vault — every past guide and master prompt
✓New drops every week

No spam — just the launch ping when the membership opens. 🩷

Step Four

Post it

Export all five at 4:5 and post them as one carousel in the calm-to-wild order. A simple 'comment CARTOON for the prompts' caption turns the saves into DMs.

Carousel order

Sleepy → Goofy → Sweet → Daydreamer → Wild. Opening soft and ending on the wildest keeps people swiping to the last slide, which the algorithm rewards.

Goal

A recognizable-but-caricatured set people tag themselves in and share — one upload, five personalities, so everyone finds 'their' one.

The read

Likeness first: if a friend can't tell it's you at a glance, dial the caricature back and re-lock the real hair/eye colour and facial identity.

Same recipe, your subject

This works on anyone — a friend, a pet owner, a founder for a brand post. Keep the five shapes and the clay/felt recipe; just feed a new photo and match the hair/eye colour. Five shareable portraits from one picture, every time.

Post it as a collab

Made a video with this? Add @ai.toon.studio_ as a collaborator when you post — we share it to our followers, and you get the reach. 🎬

9.1M views · 6 Sep 2026 — a creator built a series in this format. See who's making these →

Quality checklist

Before you hit publish

✓One clear, front-facing, shoulders-up source photo for all five
✓Hair and eye colour in each prompt changed to match the real photo
✓Facial identity stays recognizable despite the caricature — it still looks like them
✓Same handcrafted recipe on all five: matte clay skin, felt hair, soft studio light + rim light
✓Each style keeps its own head/feature shapes and solid background colour
✓All five exported 4:5, one portrait per image, full bleed, no text or watermark
✓Posted as a carousel in calm-to-wild order, with a 'comment CARTOON' caption

Creator toolstack

The tools we actually use

Built with an image model that accepts a reference photo (the same stack behind every portrait on the studio) — one upload, five full-image prompts, no video, no extra tools.

Higgsfield

Our #1 pick · animation

The fastest way to turn two key frames into a 15-second drop with real cinematic motion — camera moves, presets, no API wiring. And it's the one tool that pays you back: post through Higgsfield Earn and it pays per views.

Try Higgsfield →

Seedance 2.0

Image-to-video

Feed your two reference images plus the master prompt; renders 9:16 with consistent character identity.

Try Seedance 2.0 →

Nano Banana / GPT Image

Character refs

Lock clean, front-lit reference images for The subject so identity stays rock-solid across every beat.

Try Nano Banana / GPT Image →

ElevenLabs SFX

Cartoon sound

Generate the bzzzt · snip · click · spa "ahhh" cartoon SFX that carry the comedy — no dialogue needed.

Try ElevenLabs SFX →

Some links above are affiliate links. If you sign up through them we may earn a commission, at no extra cost to you. We only recommend tools we actually use to make our toons.

Cartoon Portraits