OpenAI CEO Sam Altman on Wednesday showcased a range of AI-generated portrait styles created using the company’s new GPT Image 1.5 model. The feature has now rolled out more widely and is being positioned as OpenAI’s answer to Google Gemini’s image generation tool, Nano Banana Pro.

Since the announcement, the official ChatGPT account on X has been teasing different portrait styles that users can try by uploading their own photos. These styles turn regular images into highly detailed illustrations, toys, sketches and cartoons, all while keeping the subject recognisable.

By combining the styles shared by Sam Altman and those posted by OpenAI, here are seven AI portrait styles users can experiment with using GPT Image 1.5.

3D Graphite pencil sketch

AI Prompt:
“Generate an image from the uploaded photo that reimagines the subject as an ultra-detailed 3D graphite pencil sketch on textured white notebook paper. Emphasise crisp paper grain, subtle imperfections, and natural surface fibres. Show the subject actively drawing, with their hand holding a pencil as the sketch comes to life. Include an eraser, sharpener, and scattered pencil shavings resting on the page. Add realistic shadows, smudges, and fine graphite residue around the working area to reinforce the tactile, hand-drawn feel.”

This style makes the image look like a lifelike pencil drawing on real paper, complete with shadows and smudges.

Classic departmental store-style portrait

AI Prompt:
“Create a classic department store-style studio portrait from the late 1980s to early 1990s. Only include the people and/or animals in the uploaded photo. Awkwardly wholesome, mall-photo-studio energy. Soft beige and grey muslin backdrop with subtle mottling. Flat, even studio lighting with mild flash glare and gentle shadows under the chin. Subject posed stiffly but earnestly, slightly turned shoulders, hands folded or resting on knees. Styling includes oversized sweaters, tucked polos, denim, pleated khakis, sensible sneakers, and conservative jewellery. Hair is voluminous, feathered, or lightly permed. Expressions are polite, mildly uncomfortable smiles. Colours are muted pastels and warm neutrals. Shot on medium-format film with soft focus, slight grain, and a faint vignette. Composition feels affordable, mass-market, and deeply nostalgic. No modern fashion, no cinematic lighting, no high contrast, no luxury styling. Authentic department-store portrait realism.”

This recreates old-school studio photos commonly found in shopping malls during the late 80s and early 90s.

K-pop idol

AI Prompt:
“Create a classic department store-style studio portrait from the late 1980s to early 1990s. Only include the people and/or animals in the uploaded photo. Awkwardly wholesome, mall-photo-studio energy. Soft beige and gray muslin backdrop with subtle mottling. Flat, even studio lighting with mild flash glare and gentle shadows under the chin. Subject posed stiffly but earnestly, slightly turned shoulders, hands folded or resting on knees. Styling includes oversized sweaters, tucked polos, denim, pleated khakis, sensible sneakers, and conservative jewellery. Hair is voluminous, feathered, or lightly permed. Expressions are polite, mildly uncomfortable smiles. Colours are muted pastels and warm neutrals. Shot on medium-format film with soft focus, slight grain, and a faint vignette. Composition feels affordable, mass-market, and deeply nostalgic. No modern fashion, no cinematic lighting, no high contrast, no luxury styling. Authentic department-store portrait realism.”

Despite the name, this prompt has been shared as part of the teaser set and focuses on a retro studio portrait look.

Chibi-style

AI Prompt:
“Transform the subjects or image into an adorable chibi-style character with a tiny body and an oversized head. If the image contains a person or multiple people, give them large, sparkling eyes, soft rounded facial features, and a cheerful expression while preserving their recognizable traits such as key facial features, hairstyle, accessories, or distinctive clothing; if the image contains an object, animal, or scene, reinterpret its most recognizable features using the same chibi proportions and simplified, cute styling. Keep the overall look short and cute, with smooth pastel shading and simplified details. Make the final image bright, expressive, and irresistibly charming, like a collectable chibi mascot.”

This style turns photos into cute, cartoon-like characters with big heads and expressive eyes.

3D Glam doll

AI Prompt:
“Transform the person in the uploaded photo into a high-end 3D glam doll with a polished, collectable toy aesthetic. Preserve the subject’s exact facial structure, skin tone, eye shape, hairstyle, and key identifying features so the person remains clearly recognisable.

Give the doll smooth porcelain-like skin with a subtle glow, softly exaggerated facial proportions, large expressive eyes with detailed lashes, and glossy lips. Add refined makeup with shimmer highlights, blush, and perfectly sculpted brows for a glamorous look.

Style the hair with realistic 3D strands, soft volume, and shine, matching the original hairstyle while enhancing it for a fashion-doll appearance. Dress the doll in a stylish, modern outfit with premium textures such as satin, leather, or metallic fabric, designed like a luxury fashion doll.

Render the image in ultra-high quality 3D with studio lighting, soft shadows, and a clean pastel or neutral background. The final result should look like a realistic, Pixar-level 3D glam doll product photo, cute yet elegant, with sharp details and a professional finish.”

This creates a luxury doll version of the person, similar to a high-end toy or animated character.

Plushie

AI Prompt:
“Transform the person in the uploaded photo into an adorable, high-quality plushie toy. Preserve the subject’s recognisable facial structure, skin tone, eye shape, hairstyle, and key identifying features so the plushie clearly resembles the original person.
Give the plushie soft fabric textures with visible stitching, fuzzy felt or velour material, and rounded, huggable proportions. Slightly exaggerate the head size while keeping the body small and cuddly. Use embroidered or button-style eyes, simplified facial details, and a warm, cheerful expression.
Recreate the hairstyle using plush fabric, yarn, or felt strands that match the original look in colour and shape. Dress the plushie in a simplified version of the original outfit using soft cloth textures, stitched seams, and gentle folds.
Render the image in ultra-high-quality 3D with soft studio lighting, gentle shadows, and a clean pastel or cosy background. The final result should look like a premium collectable plush toy product photo, cute, comforting, and highly detailed.”

This style turns people into soft, cuddly plush toys with stitched details.

Doodle

AI Prompt:
“Transform the person in the uploaded photo into a fun, expressive doodle-style illustration. Preserve the subject’s recognisable facial features, hairstyle, and overall likeness so the character clearly resembles the original person.
Use loose, hand-drawn lines with a sketchy, imperfect look, slightly exaggerated facial expressions, and simplified proportions. Add bold outlines, minimal shading, and playful details such as rosy cheeks, sparkles, or small icons around the character.
Apply flat or lightly textured colours with a bright, cheerful palette. Keep the face expressive with simple dots or oval eyes, curved brows, and a friendly smile. Recreate the hairstyle and outfit in a simplified doodle form while maintaining key identifying elements.
Render the final image as a clean digital doodle that looks like it was drawn with a pen or marker on paper, with a white or pastel background. The overall style should feel cute, casual, and lively, like a personalised cartoon doodle.”

OpenAI CEO Sam Altman on Wednesday showcased a range of AI-generated portrait styles created using the company’s new GPT Image 1.5 model. The feature has now rolled out more widely and is being positioned as OpenAI’s answer to Google Gemini’s image generation tool, Nano Banana Pro.

Since the announcement, the official ChatGPT account on X has been teasing different portrait styles that users can try by uploading their own photos. These styles turn regular images into highly detailed illustrations, toys, sketches and cartoons, all while keeping the subject recognisable.

By combining the styles shared by Sam Altman and those posted by OpenAI, here are seven AI portrait styles users can experiment with using GPT Image 1.5.

3D Graphite pencil sketch

AI Prompt:
“Generate an image from the uploaded photo that reimagines the subject as an ultra-detailed 3D graphite pencil sketch on textured white notebook paper. Emphasise crisp paper grain, subtle imperfections, and natural surface fibres. Show the subject actively drawing, with their hand holding a pencil as the sketch comes to life. Include an eraser, sharpener, and scattered pencil shavings resting on the page. Add realistic shadows, smudges, and fine graphite residue around the working area to reinforce the tactile, hand-drawn feel.”

This style makes the image look like a lifelike pencil drawing on real paper, complete with shadows and smudges.

Classic departmental store-style portrait

AI Prompt:
“Create a classic department store-style studio portrait from the late 1980s to early 1990s. Only include the people and/or animals in the uploaded photo. Awkwardly wholesome, mall-photo-studio energy. Soft beige and grey muslin backdrop with subtle mottling. Flat, even studio lighting with mild flash glare and gentle shadows under the chin. Subject posed stiffly but earnestly, slightly turned shoulders, hands folded or resting on knees. Styling includes oversized sweaters, tucked polos, denim, pleated khakis, sensible sneakers, and conservative jewellery. Hair is voluminous, feathered, or lightly permed. Expressions are polite, mildly uncomfortable smiles. Colours are muted pastels and warm neutrals. Shot on medium-format film with soft focus, slight grain, and a faint vignette. Composition feels affordable, mass-market, and deeply nostalgic. No modern fashion, no cinematic lighting, no high contrast, no luxury styling. Authentic department-store portrait realism.”

This recreates old-school studio photos commonly found in shopping malls during the late 80s and early 90s.

K-pop idol

AI Prompt:
“Create a classic department store-style studio portrait from the late 1980s to early 1990s. Only include the people and/or animals in the uploaded photo. Awkwardly wholesome, mall-photo-studio energy. Soft beige and gray muslin backdrop with subtle mottling. Flat, even studio lighting with mild flash glare and gentle shadows under the chin. Subject posed stiffly but earnestly, slightly turned shoulders, hands folded or resting on knees. Styling includes oversized sweaters, tucked polos, denim, pleated khakis, sensible sneakers, and conservative jewellery. Hair is voluminous, feathered, or lightly permed. Expressions are polite, mildly uncomfortable smiles. Colours are muted pastels and warm neutrals. Shot on medium-format film with soft focus, slight grain, and a faint vignette. Composition feels affordable, mass-market, and deeply nostalgic. No modern fashion, no cinematic lighting, no high contrast, no luxury styling. Authentic department-store portrait realism.”

Despite the name, this prompt has been shared as part of the teaser set and focuses on a retro studio portrait look.

Chibi-style

AI Prompt:
“Transform the subjects or image into an adorable chibi-style character with a tiny body and an oversized head. If the image contains a person or multiple people, give them large, sparkling eyes, soft rounded facial features, and a cheerful expression while preserving their recognizable traits such as key facial features, hairstyle, accessories, or distinctive clothing; if the image contains an object, animal, or scene, reinterpret its most recognizable features using the same chibi proportions and simplified, cute styling. Keep the overall look short and cute, with smooth pastel shading and simplified details. Make the final image bright, expressive, and irresistibly charming, like a collectable chibi mascot.”

This style turns photos into cute, cartoon-like characters with big heads and expressive eyes.

3D Glam doll

AI Prompt:
“Transform the person in the uploaded photo into a high-end 3D glam doll with a polished, collectable toy aesthetic. Preserve the subject’s exact facial structure, skin tone, eye shape, hairstyle, and key identifying features so the person remains clearly recognisable.

Give the doll smooth porcelain-like skin with a subtle glow, softly exaggerated facial proportions, large expressive eyes with detailed lashes, and glossy lips. Add refined makeup with shimmer highlights, blush, and perfectly sculpted brows for a glamorous look.

Style the hair with realistic 3D strands, soft volume, and shine, matching the original hairstyle while enhancing it for a fashion-doll appearance. Dress the doll in a stylish, modern outfit with premium textures such as satin, leather, or metallic fabric, designed like a luxury fashion doll.

Render the image in ultra-high quality 3D with studio lighting, soft shadows, and a clean pastel or neutral background. The final result should look like a realistic, Pixar-level 3D glam doll product photo, cute yet elegant, with sharp details and a professional finish.”

This creates a luxury doll version of the person, similar to a high-end toy or animated character.

Plushie

AI Prompt:
“Transform the person in the uploaded photo into an adorable, high-quality plushie toy. Preserve the subject’s recognisable facial structure, skin tone, eye shape, hairstyle, and key identifying features so the plushie clearly resembles the original person.
Give the plushie soft fabric textures with visible stitching, fuzzy felt or velour material, and rounded, huggable proportions. Slightly exaggerate the head size while keeping the body small and cuddly. Use embroidered or button-style eyes, simplified facial details, and a warm, cheerful expression.
Recreate the hairstyle using plush fabric, yarn, or felt strands that match the original look in colour and shape. Dress the plushie in a simplified version of the original outfit using soft cloth textures, stitched seams, and gentle folds.
Render the image in ultra-high-quality 3D with soft studio lighting, gentle shadows, and a clean pastel or cosy background. The final result should look like a premium collectable plush toy product photo, cute, comforting, and highly detailed.”

This style turns people into soft, cuddly plush toys with stitched details.

Doodle

AI Prompt:
“Transform the person in the uploaded photo into a fun, expressive doodle-style illustration. Preserve the subject’s recognisable facial features, hairstyle, and overall likeness so the character clearly resembles the original person.
Use loose, hand-drawn lines with a sketchy, imperfect look, slightly exaggerated facial expressions, and simplified proportions. Add bold outlines, minimal shading, and playful details such as rosy cheeks, sparkles, or small icons around the character.
Apply flat or lightly textured colours with a bright, cheerful palette. Keep the face expressive with simple dots or oval eyes, curved brows, and a friendly smile. Recreate the hairstyle and outfit in a simplified doodle form while maintaining key identifying elements.
Render the final image as a clean digital doodle that looks like it was drawn with a pen or marker on paper, with a white or pastel background. The overall style should feel cute, casual, and lively, like a personalised cartoon doodle.”