Így készült a Stílusváltó

Western 2D cartoon style, thick clean outlines, flat bright colours, squash-and-stretch expressions, TV animation look, no 3D shading, no photorealism.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

Western 2D cartoon style, thick clean outlines, flat bright colours, squash-and-stretch expressions, TV animation look, no 3D shading, no photorealism.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

Real-time video game cinematic rendered in a modern game engine, detailed PBR materials, ray-traced lighting, slightly glossy game-character look.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

Real-time video game cinematic rendered in a modern game engine, detailed PBR materials, ray-traced lighting, slightly glossy game-character look.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

Live-action, photorealistic, natural daylight from the windows, shot on a 35mm cinema camera.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

Live-action, photorealistic, natural daylight from the windows, shot on a 35mm cinema camera.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

2D anime style, cel shaded, flat colours, clean hand-inked line art, no 3D shading, no photorealism.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

2D anime style, cel shaded, flat colours, clean hand-inked line art, no 3D shading, no photorealism.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

Stylized 3D animated feature film look, soft rounded shapes, big expressive eyes, subsurface skin, warm studio lighting.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

Stylized 3D animated feature film look, soft rounded shapes, big expressive eyes, subsurface skin, warm studio lighting.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

Stop-motion claymation, handmade plasticine figures with fingerprints in the clay, miniature set, slightly jerky stop-motion movement.

In a school classroom with desks and a chalkboard, three young adult women students, all about 20 years old, crowd around a phone.
The one holding the phone wears round glasses, a messy bun with a pencil stuck in it, headphones around her neck and an oversized green knitted cardigan; she grins proudly and swipes the screen.
Next to her, a thin girl with straight black hair, a side-swept fringe over one eye, one bright red streak and heavy black eyeliner, wearing a black band t-shirt.
On the other side, a girl with long wavy silver-lavender hair and a huge pointed witch hat with a patch, in a short cape over a starry dress.
As the swipe lands, all three gasp, look down at their own hands and at each other in amazement. Static medium shot, eye level.

Stop-motion claymation, handmade plasticine figures with fingerprints in the clay, miniature set, slightly jerky stop-motion movement.

{ "duration": 4, "resolution": "480P", "aspect_ratio": "9:16", "seed": 4242, "prompt_expansion_mode": "disabled" }

Redraw this picture as a western 2D cartoon: bold black outlines, flat bright colours, simple shapes, exaggerated cartoon faces, Saturday-morning cartoon style. Keep exactly the same composition, camera angle and framing, the same three young women (about 20) in the same positions and poses, the same hand positions, the phone in the middle girl's hands, the same facial expressions and gaze directions, the same hairstyles and hair colours (left: black short messy hair with a red streak; middle: auburn hair in a messy bun with a pencil, round glasses; right: long silver-lavender hair, big navy witch hat), the same clothes and colours (left: black band tee and black jeans; middle: green cardigan and brown plaid skirt; right: navy starry dress with striped stockings), the green chalkboard behind them and the wooden desk edge in the foreground. Only the rendering style changes.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 4242, "enable_safety_checker": true }

Redraw this picture as a high quality 2D Japanese TV anime still: clean line art, cel shading, anime eyes and faces, painted anime background. Keep exactly the same composition, camera angle and framing, the same three young women (about 20) in the same positions and poses, the same hand positions, the phone in the middle girl's hands, the same facial expressions and gaze directions, the same hairstyles and hair colours (left: black short messy hair with a red streak; middle: auburn hair in a messy bun with a pencil, round glasses; right: long silver-lavender hair, big navy witch hat), the same clothes and colours (left: black band tee and black jeans; middle: green cardigan and brown plaid skirt; right: navy starry dress with striped stockings), the green chalkboard behind them and the wooden desk edge in the foreground. Only the rendering style changes.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 4242, "enable_safety_checker": true }

Re-render this picture as a still from a modern 3D animated feature film: stylized CG characters with big expressive eyes, soft subsurface skin, detailed hair and cloth, cinematic soft lighting. Keep exactly the same composition, camera angle and framing, the same three young women (about 20) in the same positions and poses, the same hand positions, the phone in the middle girl's hands, the same facial expressions and gaze directions, the same hairstyles and hair colours (left: black short messy hair with a red streak; middle: auburn hair in a messy bun with a pencil, round glasses; right: long silver-lavender hair, big navy witch hat), the same clothes and colours (left: black band tee and black jeans; middle: green cardigan and brown plaid skirt; right: navy starry dress with striped stockings), the green chalkboard behind them and the wooden desk edge in the foreground. Only the rendering style changes.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 4242, "enable_safety_checker": true }

Turn this picture into a real photograph: real young women in a real school classroom, natural skin texture, real hair and fabric, natural soft daylight, shot on a smartphone, photorealistic, no CG look. Keep exactly the same composition, camera angle and framing, the same three young women (about 20) in the same positions and poses, the same hand positions, the phone in the middle girl's hands, the same facial expressions and gaze directions, the same hairstyles and hair colours (left: black short messy hair with a red streak; middle: auburn hair in a messy bun with a pencil, round glasses; right: long silver-lavender hair, big navy witch hat), the same clothes and colours (left: black band tee and black jeans; middle: green cardigan and brown plaid skirt; right: navy starry dress with striped stockings), the green chalkboard behind them and the wooden desk edge in the foreground. Only the rendering style changes.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 4242, "enable_safety_checker": true }

Make this a real photograph of three real young women (about 20) standing in front of a chalkboard in a real classroom, as if taken with a smartphone. Real human faces with natural proportions and normal-sized eyes, real skin with pores, real hair strands, real fabric folds; natural daylight. Not a 3D render, not anime, not a doll. Keep the same composition, framing, positions, poses, hands, the phone in the middle girl's hands, and the same look of each person: left, short messy black hair with a red streak, black band tee with a heart print, black jeans, striped arm warmers; middle, auburn hair in a messy bun with a pencil in it, round glasses, green cardigan, brown plaid skirt; right, long silver-lavender hair, a big navy witch hat with a gold band, a navy dress with small stars, striped stockings. The chalkboard and the wooden desk edge in the foreground stay.

{ "image_urls": [ "⟨K-psx.png⟩" ], "aspect_ratio": "9:16", "num_images": 1, "output_format": "png" }

Turn this into an authentic candid smartphone photo of three real human young women (about 20) in a real classroom. Replace the doll-like CG faces with real human faces: natural human proportions, normal-sized human eyes, real skin texture with pores and slight imperfections, real hair strands, real fabric with wrinkles, real lighting and shadows, slight phone-camera noise. Absolutely photorealistic, like a real photo, not a 3D render, not a doll, not anime. Keep the framing, the positions and poses, the phone in the middle girl's hands, the hairstyles and hair colours, the glasses, the witch hat, all clothes and their colours, the chalkboard behind them and the desk edge in front.

{ "image_urls": [ "⟨valosag.jpg⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 777, "enable_safety_checker": true }

Make a real photograph with the exact composition of image 1: the same camera angle and framing, the chalkboard behind, the wooden desk edge in the foreground, three young women standing side by side in the same positions and poses, the middle one holding a phone at her chest with both hands, all three looking at the camera with a slight smile. The people must be the real women from image 2, photorealistic, with real human faces, skin, hair and clothes like in image 2: on the left the girl with long black hair with a red streak and dark eyeliner in a black band tee; in the middle the girl with the auburn messy bun, round glasses and the green knit cardigan; on the right the girl with long silver hair in the black witch hat. Natural classroom daylight, smartphone photo, no border.

{ "image_urls": [ "⟨K-psx.png⟩", "⟨t1-valo-frame.png⟩" ], "image_size": { "width": 1080, "height": 1920 }, "num_images": 1, "seed": 4242, "enable_safety_checker": true }

3D animated movie scene. The girl on the left with black hair and a red streak looks at herself, grins and does a pleased little pose; the others look amazed at their shiny new look. Smooth expressive 3D animation. Static camera, no cuts. At the end they are back in exactly the starting pose.

{ "image_url": "⟨3d-anim.jpg⟩", "end_image_url": "⟨3d-anim.jpg⟩", "duration": 2.4, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Western 2D cartoon scene. The witch girl on the right looks down at herself and her jaw drops in an exaggerated cartoon way, eyes popping; the other two look at her and giggle. Bouncy cartoon animation. Static camera, no cuts. At the end they are back in exactly the starting pose.

{ "image_url": "⟨rajzfilm.jpg⟩", "end_image_url": "⟨rajzfilm.jpg⟩", "duration": 2.4, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

2D anime scene. The three girls suddenly realise they have turned into anime characters: the girl on the left gasps with huge sparkling eyes, the girl in the middle with glasses talks excitedly and proudly shows her phone, the witch girl on the right looks at her own hands in amazement. Lively anime acting, hair moving. Static camera, no cuts. At the end they are back in exactly the starting pose.

{ "image_url": "⟨anime.jpg⟩", "end_image_url": "⟨anime.jpg⟩", "duration": 4.6, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Turn image 1 into a real candid smartphone photo taken in a real school classroom. These are real adult women in their twenties with real human faces, real proportions, natural skin texture, real hair and real fabric: not anime, not 3D, not a doll, not a costume render. Keep only the layout of image 1: the camera angle, the green chalkboard behind, the wooden desk edge in the foreground, and the three women standing side by side in the same places. Left: emo girl, black choppy hair with one red streak, black band t-shirt, black jeans, striped red-black arm warmers, studded belt. Middle: nerdy girl with an auburn messy bun with a pencil in it, round glasses, oversized green knitted cardigan, brown plaid skirt, holding a phone to her chest. Right: witch-costume girl with long silver-lavender wavy hair, huge navy pointed witch hat, navy dress with gold stars, striped stockings. New in this frame: a fourth woman, the one in image 2, walks in from the right edge mid-stride, half in frame, turning her head toward the camera with a dramatic, offended pout: very long straight black hair with side-swept bangs, flowing pale lavender maxi dress. The three turn to look at her, surprised. Natural daylight from the windows, slight phone-camera noise.

{ "image_urls": [ "⟨K-psx-ref.png⟩", "⟨cepi-ref.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "low", "num_images": 1, "output_format": "png" }

Keep the exact layout of image 1, because this frame will be cut against other versions of the same shot: the same camera position, angle and framing, the same three women at exactly the same places in the frame, the same sizes, with their heads, shoulders, hands and the desk edge at the same positions. Same poses: the middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them the same green chalkboard, completely blank, no writing; in front the same wooden desk edge filling the bottom part of the frame. No text, letters or numbers anywhere in the picture; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, studded belt, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil stuck in it, round glasses, oversized green knitted cardigan, brown plaid skirt with mustard lines, dark tights. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat with a gold band, navy dress with small gold stars, short cape, navy-white striped stockings. All three are adult women in their twenties. STYLE: a frame from a modern big-budget 3D animated feature film: stylized CG characters with big expressive eyes, soft subsurface skin, detailed groomed hair, rich cloth, warm cinematic lighting with soft bounce light and a gentle depth of field on the background. Polished and high-end; nothing of the low-poly PS1 look of image 1 remains, only its layout.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Keep the exact layout of image 1, because this frame will be cut against other versions of the same shot: the same camera position, angle and framing, the same three women at exactly the same places in the frame, the same sizes, with their heads, shoulders, hands and the desk edge at the same positions. Same poses: the middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them the same green chalkboard, completely blank, no writing; in front the same wooden desk edge filling the bottom part of the frame. No text, letters or numbers anywhere in the picture; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, studded belt, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil stuck in it, round glasses, oversized green knitted cardigan, brown plaid skirt with mustard lines, dark tights. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat with a gold band, navy dress with small gold stars, short cape, navy-white striped stockings. All three are adult women in their twenties. STYLE: a real candid smartphone photo in a real school classroom. Real people with real human faces and proportions, natural skin texture, real hair and fabric, natural window daylight, slight phone-camera noise. Not anime, not 3D, not a doll, not a render. Every face, body and material is fully photographic; only the layout comes from image 1.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Keep the exact layout of image 1, because this frame will be cut against other versions of the same shot: the same camera position, angle and framing, the same three women at exactly the same places in the frame, the same sizes, with their heads, shoulders, hands and the desk edge at the same positions. Same poses: the middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them the same green chalkboard, completely blank, no writing; in front the same wooden desk edge filling the bottom part of the frame. No text, letters or numbers anywhere in the picture; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, studded belt, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil stuck in it, round glasses, oversized green knitted cardigan, brown plaid skirt with mustard lines, dark tights. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat with a gold band, navy dress with small gold stars, short cape, navy-white striped stockings. All three are adult women in their twenties. STYLE: a frame from an original western 2D TV cartoon: thick black outlines, flat bright colours, simple rubbery shapes, exaggerated cartoon proportions with big heads and big expressive cartoon faces, a simple flat-coloured background. Fully 2D cartoon; nothing of the low-poly 3D look of image 1 remains, only its layout.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Keep the exact layout of image 1, because this frame will be cut against other versions of the same shot: the same camera position, angle and framing, the same three women at exactly the same places in the frame, the same sizes, with their heads, shoulders, hands and the desk edge at the same positions. Same poses: the middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them the same green chalkboard, completely blank, no writing; in front the same wooden desk edge filling the bottom part of the frame. No text, letters or numbers anywhere in the picture; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, studded belt, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil stuck in it, round glasses, oversized green knitted cardigan, brown plaid skirt with mustard lines, dark tights. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat with a gold band, navy dress with small gold stars, short cape, navy-white striped stockings. All three are adult women in their twenties. STYLE: a keyframe from a modern Japanese TV anime: clean confident line art, cel shading with two tones, big expressive anime eyes, anime hair with sharp highlights, a painted anime classroom background with soft light. Fully hand-drawn 2D; nothing of the low-poly 3D look of image 1 remains, only its layout.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Keep from image 1 only these things, because this frame will be cut against other versions of the same shot: the camera angle and framing, where each of the three women stands in the frame and how much of the frame she fills, her pose, her hair colour and hairstyle and her outfit with its colours. The middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them a blank green chalkboard with no writing, in front a wooden desk edge filling the bottom part of the frame. Redesign everything else from scratch, as a character designer of the style below would draw these three women: new faces, new head and body proportions, new hands. They must not look like the low-poly anime 3D models of image 1. No text, letters or numbers anywhere; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil in it, round glasses, green cardigan, brown plaid skirt. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat, navy dress with gold stars, navy-white striped stockings. All three are adult women in their twenties. STYLE: an original modern western 2D TV cartoon: thick slightly wobbly black outlines, flat colours with no gradients, very simple geometric shapes, big heads, thin noodle-like arms and legs, simple oval eyes with small pupils, big expressive mouths, chunky simplified hair shapes, a flat simple background. Clearly a goofy western cartoon, not anime and not manga.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Keep from image 1 only these things, because this frame will be cut against other versions of the same shot: the camera angle and framing, where each of the three women stands in the frame and how much of the frame she fills, her pose, her hair colour and hairstyle and her outfit with its colours. The middle woman holds her phone against her chest with its blank, softly glowing screen facing the camera; all three look into the camera with a small smile. Behind them a blank green chalkboard with no writing, in front a wooden desk edge filling the bottom part of the frame. Redesign everything else from scratch, as a character designer of the style below would draw these three women: new faces, new head and body proportions, new hands. They must not look like the low-poly anime 3D models of image 1. No text, letters or numbers anywhere; the black t-shirt only has a winged broken heart graphic. Left: emo woman, black choppy hair with one red streak, black choker, black t-shirt, black jeans, red-black striped arm warmers. Middle: nerdy woman, auburn messy bun with a pencil in it, round glasses, green cardigan, brown plaid skirt. Right: witch woman, long wavy silver-lavender hair, huge navy pointed witch hat, navy dress with gold stars, navy-white striped stockings. All three are adult women in their twenties. STYLE: a frame from a modern 3D animated family feature film: appealing stylized CG designs with rounded shapes, larger heads, big glossy eyes, small noses, soft subsurface skin, sculpted stylized hair, rich cloth textures, warm cinematic lighting with soft bounce light and a gentle depth of field. Clearly a big-budget animated movie character design, not anime and not a game model.

{ "image_urls": [ "⟨K-psx.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Image 1 is a frame from a 3D animated feature film. Keep image 1 exactly as it is: same camera, framing, lighting, background, desk, and the same three women unchanged in the same places and poses. Add one more character: the woman of image 2, redesigned in exactly the same 3D animated film style as image 1 (not low-poly, not anime). She has just walked in from the right and now stands at the right edge, slightly in front of the witch, fully visible from her head down to the desk edge, at the same height and scale as the others. Her body is turned three quarters toward the middle woman's phone, one hand on her chest, and she gazes at the phone screen with a delighted, enchanted, slightly vain expression, as if she has just seen the most beautiful version of herself. She is the most glamorous of them, like an animated film princess: very long straight glossy black hair down to her waist with side-swept bangs, big violet eyes, a flowing pale lavender maxi dress with a thin belt and small white flower embroidery near the hem. An adult woman in her twenties. No text, letters or numbers anywhere.

{ "image_urls": [ "⟨kf-3d2.png⟩", "⟨cepi-ref.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "low", "reasoning_effort": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, the same head, hair, face, neck, shoulders and arms at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is a CHARACTER AND STYLE reference only (her hair, colours, clothes and the drawing style); ignore its composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical close-up, eye level, frontal): one young woman faces the camera and looks straight into the lens. Her spiky hair fills the top of the frame and is cut off by the top edge; the hair spreads from about 8% to 90% of the frame width. Her face spans about 29% to 71% of the width; her eyes sit at about 31-37% of the height, her mouth at about 46%, her chin at about 52%. Her neck with a choker is at 52-60% of the height. Her shoulders and upper chest fill the bottom 40% of the frame and are cut off by the bottom edge; both upper arms hang down at her sides and are cut off by the left and right edges. Her mouth is closed, with a small calm smile. Nothing else is in the frame; there are no hands visible.

The woman is the character from image 2: short messy black hair with one red streak at the front, large blue-grey anime eyes, a black choker, a black band T-shirt.

Style: a 2D Japanese TV anime still, clean confident line art, flat cel shading with one shadow tone, soft even daylight, no dramatic lighting. Behind her a plain green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the hair outline, face, chin, neck or shoulders.

{ "image_urls": [ "⟨agyag-eleje.png⟩", "⟨emoka-anime-ref.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, the same head, hair, face, neck, shoulders and arms at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is a CHARACTER AND STYLE reference only (her hair, colours, clothes and the drawing style); ignore its composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical close-up, eye level, frontal): one young woman faces the camera and looks straight into the lens. Her spiky hair fills the top of the frame and is cut off by the top edge; the hair spreads from about 8% to 90% of the frame width. Her face spans about 29% to 71% of the width; her eyes sit at about 31-37% of the height, her mouth at about 46%, her chin at about 52%. Her neck with a choker is at 52-60% of the height. Her shoulders and upper chest fill the bottom 40% of the frame and are cut off by the bottom edge; both upper arms hang down at her sides and are cut off by the left and right edges. Her mouth is closed, with a small calm smile. Nothing else is in the frame; there are no hands visible.

The woman is the character from image 2: short messy black hair with one red streak at the front, large blue-grey anime eyes, a black choker, a black band T-shirt.

Style: a 2D Japanese TV anime still, clean confident line art, flat cel shading with one shadow tone, soft even daylight, no dramatic lighting. Behind her a plain green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the hair outline, face, chin, neck or shoulders.

{ "image_urls": [ "⟨agyag-eleje.png⟩", "⟨emoka-anime-ref.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "num_images": 1, "output_format": "png", "enhance_prompt_mode": "standard" }

Edit image 1. Change ONLY her pose and facial expression, to match the pose in image 2: she raises both hands and presses her open palms against her cheeks, one hand on each side of her face, fingers pointing up, the forearms rising from the bottom edge of the frame; her mouth is open in a small round O of shock, her eyebrows are raised, her eyes are wide open and still look into the lens. Keep everything else from image 1 exactly as it is: the same framing and camera, the same head size and position, the same hair, choker, T-shirt, colours, green chalkboard background and the same anime line art and flat cel shading. Image 2 is a grey 3D pose guide only: use it for the position of the hands and the open mouth, never for colours, shapes of the hair or style. Do not reframe, recenter or zoom. Check: overlaid at 50% on image 1, the hair, face outline, neck and shoulders have no double edges; only the hands and the mouth differ.

{ "image_urls": [ "⟨kezdo.png⟩", "⟨agyag-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Stylized 2D Japanese TV anime shot, one continuous take, static locked camera, 9:16 close-up. The first frame is image 1, the last frame is the end image: the same young woman, same framing. 0.0-0.4 s: she looks into the lens with a small calm smile, still. 0.4-1.1 s: she says a short shocked line, her lips move with the voice. 1.1-2.5 s: her eyes go wide, her eyebrows shoot up, she raises both hands and presses her palms to her cheeks, her mouth stays open in a small round O. Keep her hair, face, clothes, colours and the flat cel-shaded line art exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨kezdo.png⟩", "end_image_url": "⟨zaro.png⟩", "target_audio_url": "⟨emoka-e1-3s.wav⟩", "duration": 2.5, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Video 1 is the motion and camera reference only: a grey 3D blockout of the shot; copy its timing and movement exactly (the head, the two hands rising to the cheeks, the mouth opening), static camera, same framing. Image 1 is the look: the first frame of the shot, a stylized 2D Japanese TV anime close-up of this young woman; the whole video must look exactly like image 1 (her hair, face, clothes, colours, green chalkboard background, flat cel-shaded line art) and never like the grey blockout. 0.0-0.4 s: calm small smile into the lens. 0.4-1.1 s: she says a short shocked line, lips moving with the voice. 1.1-2.8 s: eyes wide, eyebrows up, both palms pressed to her cheeks, mouth in a small round O. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text, no grey 3D look.

{ "reference_video_urls": [ "⟨agyag-2-lassu.mp4⟩" ], "reference_image_urls": [ "⟨kezdo.png⟩" ], "target_audio_url": "⟨emoka-e1-3s.wav⟩", "duration": 2.8, "resolution": "480P", "aspect_ratio": "9:16", "prompt_expansion_mode": "disabled", "seed": 4242 }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is a CHARACTER AND STYLE reference only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical medium close-up, eye level, frontal): Kocka stands in the middle and looks into the lens with a small smile; her hair bun touches the top edge; her face spans about 27% to 68% of the frame width, her glasses sit at about 26% of the height, her chin at about 40%. Her torso fills the lower half of the frame and is cut off by the bottom edge. With her left hand she holds a phone at chest height on the right side of the frame (about 60% to 82% of the width, 50% to 75% of the height), its screen facing the camera. At the far left edge only a thin slice of Emóka's head and shoulder is visible, cut off by the left edge. Behind Kocka on the right, Boszi's witch hat brim and part of her face are visible, cut off by the top and right edges.

The people, as in the references: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, a green cardigan over a cream top, a brown plaid skirt; Emóka: short messy black hair with one red streak, a black choker, a black band T-shirt with a pixel heart with wings, red and black striped arm warmers; Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: photorealistic, a candid photo of real young women taken on a modern phone, soft even daylight from the classroom windows, real skin texture and real fabrics, natural colours, a slight depth of field. Behind them a green school chalkboard with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨1-valo-kocka-eleje.png⟩", "⟨valo-csoport.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is a CHARACTER AND STYLE reference only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical medium close-up, eye level, frontal): Boszi stands in the middle; the brim of her big witch hat spans almost the whole frame width at about 8% to 25% of the height, the hat crown is cut off by the top edge. Her face spans about 32% to 65% of the width, her eyes at about 42% of the height, her chin at about 55%. Her long hair falls down both sides of her body. She holds both hands up in front of her chest with the palms up, at about 50% to 80% of the height, looking at them. At the left edge a second person's hands and arm, holding a phone, are cut off by the left edge.

The people, as in the references: Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings; Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, a green cardigan over a cream top, a brown plaid skirt. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D TV cartoon still, thick clean black outlines, flat bright colours with no gradients, simple rounded shapes, big expressive cartoon eyes, soft even light. Behind them a flat green chalkboard with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨4-rajz-boszi-eleje.png⟩", "⟨rajz-csoport.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is a CHARACTER AND STYLE reference only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical medium close-up, eye level, frontal): Kocka stands in the middle, her eyes closed in a happy smile; her hair bun touches the top edge; her face spans about 30% to 68% of the frame width, her glasses at about 24% of the height, her chin at about 38%. With her right hand she holds a phone up at the left side of the frame next to her head (about 0% to 10% of the width, 30% to 50% of the height), its screen facing the camera. Her torso fills the lower half and is cut off by the bottom edge. Behind her on the right, Boszi's witch hat brim and long hair are visible, cut off by the top and right edges; behind the phone on the left a little of Emóka's black hair.

The people, as in the references: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, a green cardigan over a cream top, a brown plaid skirt; Emóka: short messy black hair with one red streak, a black choker, a black band T-shirt with a pixel heart with wings, red and black striped arm warmers; Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D Japanese TV anime still, clean confident line art, flat cel shading with one shadow tone, soft even daylight, no dramatic lighting. Behind them a plain green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨3-anime-kocka-eleje.png⟩", "⟨anime-csoport.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Images 2-3 are CHARACTER AND STYLE references only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Cépi stands in the right half of the frame, walking in toward the left; her face spans about 65% to 90% of the width at about 15% to 35% of the height; her long straight hair falls to her waist; her arms hang loosely, her right hand at about 35% to 45% of the width near the bottom. Her body is cut off by the right and bottom edges. At the left edge Boszi's hat brim and a strip of her hair and dress are visible, cut off by the left edge. The middle of the frame is the empty classroom wall and the chalkboard.

The people, as in the references: Cépi: long straight glossy black hair down to her waist with side-swept bangs, violet eyes, a light lavender maxi dress; Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨5-3d-cepi-be-eleje.png⟩", "⟨3d-csoport.png⟩", "⟨cepi-3d.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Images 2-3 are CHARACTER AND STYLE references only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical medium close-up, eye level, frontal): Kocka stands in the middle and looks into the lens with a cheeky smile; her hair bun touches the top edge; her face spans about 30% to 70% of the width, her glasses at about 28% of the height, her chin at about 42%. Her torso fills the lower half and is cut off by the bottom edge. With her left hand she holds a phone at the lower right (about 65% to 90% of the width, 60% to 95% of the height), its screen facing the camera, her right hand reaching toward it. At the top right edge a little of Boszi's hat brim and hair.

The people, as in the references: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, a green cardigan over a cream top, a brown plaid skirt; Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨7-3d-kocka-eleje.png⟩", "⟨kocka-3d.png⟩", "⟨3d-csoport.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and the phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Images 2-3 are CHARACTER AND STYLE references only (who is who, their hair, colours, clothes and the drawing style); ignore their composition and framing entirely.

Spatial survey of image 1 (a 9:16 vertical close-up, eye level): Cépi's face is big, in the middle-right of the frame, spanning about 35% to 85% of the width, her eyes at about 28% of the height, her mouth at about 38%; she looks down at the phone in her hand. Her long straight hair frames her face and falls past the bottom edge. In her right hand she holds a black phone at the lower left of the frame (about 5% to 45% of the width, 45% to 75% of the height), its back toward the camera, so she sees its screen. Behind her on the left, Boszi's witch hat and face peek in, cut off by the left edge.

The people, as in the references: Cépi: long straight glossy black hair down to her waist with side-swept bangs, violet eyes, a light lavender maxi dress; Boszi: long wavy silver-lavender hair, a big navy witch hat with a gold band, a navy strapless dress with gold stars and gold trim, striped stockings. The phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard, softly out of focus, with no writing on it.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, the phone or the shoulders.

{ "image_urls": [ "⟨6-3d-cepi-kozeli-eleje.png⟩", "⟨cepi-3d.png⟩", "⟨3d-csoport.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY this, to match image 2: Kocka lifts the phone slightly higher and her proud grin gets wider. Everything else stays exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨3-anime-kocka-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY Boszi's face, to match image 2: her eyes go wide, her eyebrows shoot up and her mouth opens in a small round O of surprise; she now looks into the lens. Everything else stays exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨4-rajz-boszi-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY this, to match image 2: Cépi has stepped a little further to the left, toward the middle of the frame, and frowns at the others with raised eyebrows, her mouth slightly open as she speaks. Keep her look, hair and dress, and everything else, exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨5-3d-cepi-be-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY this, to match image 2: Cépi, enchanted by her own reflection, raises her other hand to her cheek, her eyes shine and her lips part in awe; she still looks at the phone. Everything else stays exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨6-3d-cepi-kozeli-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY this, to match image 2: Kocka taps the phone screen with her right index finger and grins mischievously, eyebrows raised. Everything else stays exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨7-3d-kocka-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Edit image 1. Change ONLY Kocka's expression, to match image 2: her eyes squeeze shut in a big delighted grin, her cheeks pushed up. Everything else stays exactly as in image 1. Image 2 is a grey 3D pose guide only: use it for the pose and the expression, never for colours, shapes of the hair or the style. Keep the same framing and camera, the same people, hair, clothes, colours, background and the same style. Do not reframe, recenter or zoom.

Check: overlaid at 50% on image 1, only the changed parts differ; there are no double edges anywhere else.

{ "image_urls": [ "⟨eleje.png⟩", "⟨1-valo-kocka-vege.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Stylized 3D animated feature film shot, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.6 s: Cépi walks in from the right with a graceful step toward the middle, her long hair swinging. 0.6-3.2 s: she stops, frowns at the others and says an annoyed line, her lips move with the voice. Boszi at the left edge stays still and silent. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 3.5, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 2D TV cartoon shot, thick outlines and flat colours, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.3 s: Boszi looks down at her own hands, puzzled. 0.35-1.9 s: she says a short astonished line, her lips move with the voice; her eyes go wide and her eyebrows shoot up, and she looks up into the lens. The hands holding the phone at the left edge stay still. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 2.6, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 3D animated feature film shot, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.3 s: Kocka looks into the lens with a cheeky smile. 0.3-2.3 s: she says a short mischievous line, her lips move with the voice, and taps her phone screen with her finger, eyebrows raised. Boszi at the edge stays still and silent. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 2.6, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 2D Japanese TV anime shot, flat cel-shaded line art, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.3 s: Kocka grins, eyes closed, holding her phone up next to her head with the screen toward the camera. 0.35-3.6 s: she proudly says a long excited sentence, her lips move with the voice, she lifts the phone a little higher. The others behind her stay still and silent. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 3.9, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Photorealistic live-action shot, a real phone video, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.3 s: Kocka smiles into the lens, holding up her phone. 0.35-1.2 s: Kocka says a short excited line, her lips move with the voice. 1.5-2.5 s: a bored voice answers from off-screen on the left; Kocka keeps quiet and her face lights up into a big delighted grin, eyes squeezed shut. Only Kocka talks on screen; Emóka and Boszi at the edges stay still and silent. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 3, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 3D animated feature film shot, one continuous take, static locked camera, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people, the same framing. 0.0-0.2 s: Cépi stares at her reflection in the phone screen. 0.2-4.4 s: in awe she says a slow enchanted line, her lips move with the voice, a little gasp at the start; she raises her other hand to her cheek. Boszi behind her stays still and silent. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No camera movement, no zoom, no cuts, no soft dissolves or fluid morphs, no extra people, no text.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 4.7, "resolution": "480P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Kocka stands left of centre and laughs toward the lens, her eyes squeezed shut in a big grin, mouth open; her bun top at about 17% of the height, her face spans about 30% to 60% of the width, her eyes at about 32%, her chin at about 40%. In her right hand she holds a phone at chest height, about 60% to 73% of the width and 52% to 65% of the height, its screen facing the camera. Her torso is cut off by the bottom edge. On the right, Boszi is seen from the side, about 70% to 100% of the width: her arm is raised high, her hand at about 88% of the width and 12% of the height adjusting the brim of her witch hat; her long hair and navy dress, cut off by the right edge. In the bottom left corner Emóka's forearm with a striped arm warmer reaches in from the left edge at about 85% of the height.

The people: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim; Emóka: short messy black hair with one red streak, blue eyes, a black choker, a black T-shirt with a red pixel heart with white wings and white pixel lettering, black trousers with a belt, red and black striped arm warmers. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: photorealistic, a candid photo of real young women taken on a modern phone, soft even daylight from the classroom windows, real skin texture and real fabrics, natural colours, a slight depth of field. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨valo-eleje.png⟩", "⟨valo-eleje-szin.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Kocka stands left of centre and smiles into the lens; her bun top at about 17% of the height, her face spans about 30% to 55% of the width, her eyes at about 37%, chin at about 43%. In her right hand she holds a black phone at her chest, about 35% to 46% of the width and 55% to 66% of the height. Her body is cut off by the bottom edge. At the right edge a thin sliver of Boszi's hat brim and navy dress, cut off by the right edge.

The people: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D TV cartoon still, thick clean black outlines, flat bright colours with no gradients, simple rounded shapes, big expressive cartoon eyes, soft even light. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨rajz-eleje.png⟩", "⟨rajz-eleje-szin.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): on the left Emóka, about 0% to 45% of the width, cut off by the left and bottom edges; her face spans about 5% to 38% of the width, her hair top at about 17% of the height, her eyes at about 32%, a small smirk toward the lens. In the middle Kocka, about 45% to 90% of the width: her bun top at about 20% of the height, her face spans about 55% to 78% of the width, her eyes closed in a happy smile at about 38%; she holds a black phone up beside her head at about 49% to 58% of the width, 33% to 48% of the height, her arm raised. Behind Kocka on the right, Boszi: her witch hat brim spans about 72% to 100% of the width at about 20% of the height, the hat tip near the top edge; her face at about 80% to 95% of the width, her eyes at about 26%, she looks down at Kocka; cut off by the right edge.

The people: Emóka: short messy black hair with one red streak, blue eyes, a black choker, a black T-shirt with a red pixel heart with white wings and white pixel lettering, black trousers with a belt, red and black striped arm warmers; Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D Japanese TV anime still, clean confident line art, flat cel shading with one shadow tone, soft even daylight, no dramatic lighting. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨anime-eleje.png⟩", "⟨anime-eleje-szin.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): on the left Boszi, about 0% to 45% of the width, cut off by the left and bottom edges; her hat brim spans about 0% to 45% at about 18% to 30% of the height; her face spans about 5% to 30% of the width, her eyes at about 30%; she looks toward Cépi. Right of centre Cépi, about 45% to 100% of the width: her face spans about 62% to 83% of the width, her eyes at about 32% of the height, her lips slightly parted; her long straight black hair falls to about 65% of the height; her arms hang, one hand at about 42% of the width and 83% of the height. Cut off by the right and bottom edges.

The people: Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim; Cépi: long straight glossy black hair down to her waist with side-swept bangs, violet eyes, a light lavender sleeveless maxi dress. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨3d-eleje.png⟩", "⟨3d-eleje-szin.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely. Image 3 is the STYLE reference: the same people a few seconds earlier. Match its drawing style, rendering, lighting and character designs exactly; ignore its composition.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Emóka stands in the middle-left, about 15% to 62% of the width; her face spans about 23% to 54% of the width, her eyes at about 40% of the height, partly covered by Boszi's raised arm; one striped forearm crosses her chest at about 50% to 60% of the height, the other arm hangs down; cut off by the bottom edge. On the right Boszi, about 65% to 100% of the width: her eyes are closed at about 35% of the height; her witch hat brim spans about 55% to 100% of the width at about 25% to 32%, the crown cut off by the top edge. Her right arm is raised across the frame, elbow at about 40% of the width and 40% of the height in front of Emóka's face, her hand at the hat brim at about 68% of the width and 18% of the height, pushing the hat up. Cut off by the right and bottom edges.

The people: Emóka: short messy black hair with one red streak, blue eyes, a black choker, a black T-shirt with a red pixel heart with white wings and white pixel lettering, black trousers with a belt, red and black striped arm warmers; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D TV cartoon still, thick clean black outlines, flat bright colours with no gradients, simple rounded shapes, big expressive cartoon eyes, soft even light. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨rajz-vege.png⟩", "⟨rajz-vege-szin.png⟩", "⟨eleje.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely. Image 3 is the STYLE reference: the same people a few seconds earlier. Match its drawing style, rendering, lighting and character designs exactly; ignore its composition.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): at the left edge Emóka, partly hidden, about 0% to 20% of the width: her face spans about 0% to 18%, her eyes at about 31% of the height, a smirk; her lower body is hidden behind a desk; cut off by the left edge. Right of centre Kocka, about 38% to 100% of the width: her bun top at about 19% of the height, her face spans about 50% to 75% of the width, her glasses at about 35%, her mouth open as she talks toward the lens; in her right hand a black phone at about 40% to 55% of the width and 55% to 65% of the height. She is cut off by the right and bottom edges. A wooden desk behind them in the middle.

The people: Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt; Emóka: short messy black hair with one red streak, blue eyes, a black choker, a black T-shirt with a red pixel heart with white wings and white pixel lettering, black trousers with a belt, red and black striped arm warmers. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨3d-vege.png⟩", "⟨3d-vege-szin.png⟩", "⟨eleje.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely. Image 3 is the STYLE reference: the same people a few seconds earlier. Match its drawing style, rendering, lighting and character designs exactly; ignore its composition.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Emóka stands left of centre, about 0% to 50% of the width, and looks into the lens, eyes wide, her mouth a small open O; her face spans about 20% to 42% of the width, her hair top at about 21% of the height, her eyes at about 34%, her chin at about 39%. One hand touches her chest; her belt is at about 78% of the height and the bottom edge cuts her legs. On the right Kocka, about 66% to 100% of the width: her face spans about 70% to 92%, her glasses at about 36% of the height, a small smile toward the lens; she holds a black phone at her chest, about 70% to 80% of the width, 57% to 67% of the height. She is cut off by the right and bottom edges. At the top right corner a bit of Boszi's navy hat and face, cut off by the right edge.

The people: Emóka: short messy black hair with one red streak, blue eyes, a black choker, a black T-shirt with a red pixel heart with white wings and white pixel lettering, black trousers with a belt, red and black striped arm warmers; Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: photorealistic, a candid photo of real young women taken on a modern phone, soft even daylight from the classroom windows, real skin texture and real fabrics, natural colours, a slight depth of field. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨valo-vege.png⟩", "⟨valo-vege-szin.png⟩", "⟨eleje.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely. Image 3 is the STYLE reference: the same people a few seconds earlier. Match its drawing style, rendering, lighting and character designs exactly; ignore its composition.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): Boszi fills the middle. Her witch hat crown is cut off by the top edge; the brim spans about 2% to 85% of the width at about 25% to 33% of the height. Her face spans about 29% to 50% of the width, her eyes wide at about 34% of the height, her mouth a small open O at about 40%, chin at 44%. Her long lavender hair falls down both sides to about 75% of the height. Her right hand points at her own chest at about 25% to 45% of the width, 50% to 60% of the height. Her navy dress is cut off by the bottom edge. Behind her on the right, a little of Kocka's auburn hair and green sleeve at about 80% to 90% of the width.

The people: Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim; Kocka: auburn hair in a messy bun with a pencil stuck in it, round glasses, green eyes, small yellow headphones around her neck, a green striped cardigan over a cream top, a brown and yellow plaid skirt. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a 2D Japanese TV anime still, clean confident line art, flat cel shading with one shadow tone, soft even daylight, no dramatic lighting. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨anime-vege.png⟩", "⟨anime-vege-szin.png⟩", "⟨eleje.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Photorealistic live-action footage, a real phone video of real young women, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.4 s: Kocka grins and holds her phone out toward the others. 0.5-1.3 s: Kocka says a short excited line, her lips move with the voice. Meanwhile Boszi on the right adjusts the brim of her witch hat, then rises on tiptoe to peek at the phone screen. 1.3-1.7 s: Kocka pushes the phone toward Emóka. 1.7-2.7 s: Emóka pulls a small earbud out of her ear, leans forward with a bored face and says a short flat line, her lips move with the voice; Kocka keeps quiet. 3.3 s: Kocka swipes her thumb across the phone screen. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 4.2, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 2D Japanese TV anime footage, flat cel-shaded line art, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.8 s: Kocka holds her phone up beside her head, eyes closed, smiling. 0.8-1.4 s: Emóka jumps back, both hands to her cheeks, and gasps a short line, her lips move with the voice. 1.4-1.8 s: Emóka touches her own huge anime eye with one finger. 1.8-5.1 s: Kocka proudly lifts her phone high above her head and says a long excited sentence, her lips move with the voice. Meanwhile Boszi's witch hat sparkles; she spins once around to look at herself, her long hair flutters, and she ends pointing at her own chest, amazed. Only one person talks at a time. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 5.8, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 3D animated feature film footage, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.9 s: Cépi stops and looks at her sparkling hand in wonder. 0.9-5.1 s: Cépi pulls out her own phone, holds it at arm's length in selfie mode and says an enchanted line in awe, her lips move with the voice; she puts her other hand to her cheek and changes pose twice. Meanwhile Boszi peeks into the selfie over Cépi's shoulder and fixes her own hair, and Emóka rolls her eyes, folds her arms and looks at Kocka. 5.5-7.5 s: Kocka grins slyly, says a short mischievous line, her lips move with the voice, and taps her phone screen with one finger. Only one person talks at a time. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 7.6, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 2D TV cartoon footage, thick black outlines and flat bright colours, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.9 s: Kocka smiles into the lens holding her phone at her chest. 0.9-2.4 s: off to the right Boszi stares at her own four-fingered cartoon hands and says a short astonished line, her eyes grow huge and round, her lips move with the voice. 1.5-2.5 s: Kocka giggles with a hand over her mouth and swipes her phone at 2.5 s. 2.6-3.9 s: Emóka pokes the tip of Boszi's witch hat with one finger; the hat flops down over Boszi's eyes and Boszi pushes it up with one arm. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 3.9, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

A moving oil painting: thick visible brush strokes and impasto texture on everything, the paint strokes gently shimmer, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.4 s: the group stands together; Boszi pushes her witch hat back up with both hands. 0.4-2.6 s: Emóka lifts her striped arm-warmer sleeve, stares at the painted brush strokes on it and says a short puzzled line, her lips move with the voice. 2.6-4.9 s: Emóka sniffs her sleeve and says another short dry line, her lips move with the voice. Boszi folds her hands and smiles calmly. Kocka nods slowly like a visitor in a museum, and at 5.1 s swipes her phone. Only Emóka talks. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 6.2, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

A moving cubist painting in the style of early 20th century cubism: faces with both eyes on one side and the nose in profile, broken angular planes, the whole world gently wobbles and breathes, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.9 s: Cépi walks in from the right toward the group, her long black hair swinging. 0.9-3.5 s: Cépi stops, puts her hands on her hips and asks a suspicious question, her lips move with the voice. Boszi turns toward her and waves apologetically. Kocka hides her phone behind her back, then pulls it out again with a guilty grin. Only Cépi talks. The floor, the board and the walls wobble gently in angular cubist planes. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 4, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Image 1 is the ONLY source of truth for geometry and composition. Reproduce it exactly: same camera, same framing, every person, head, hand and phone at the same position, size and silhouette. Repaint surfaces only. Do not reframe, recenter or zoom. Image 2 is the same frame in a crude low-poly colour render: use it ONLY for who is who and the colours and patterns of hair, eyes, clothes and skin; ignore its blocky low-poly look entirely. Image 3 is the STYLE reference: the same people a few seconds earlier. Match its drawing style, rendering, lighting and character designs exactly; ignore its composition.

Spatial survey of image 1 (a 9:16 vertical medium shot, eye level): on the left Boszi, about 0% to 55% of the width, peeking over Cépi's shoulder; her witch hat tip at about 33% of the width and 11% of the height, the brim spans about 0% to 60% of the width at about 20% to 33%; her face spans about 27% to 48% of the width, her eyes closed in a happy smile at about 30% of the height; her long lavender hair falls to about 55%; her navy dress and striped stockings are cut off by the bottom and left edges. On the right Cépi, about 45% to 100% of the width, taking a selfie: her face spans about 57% to 83% of the width, her eyes closed in bliss at about 30% of the height, a small dreamy smile at about 35%; her long straight black hair falls to about 70%; one hand rests against her cheek at about 47% to 60% of the width, 30% to 42% of the height; her other arm is raised high at the right edge, holding a black phone up at about 84% to 97% of the width and 13% to 28% of the height, its back toward the camera and its screen toward her face. Her lavender dress is cut off by the bottom edge.

The people: Cépi: long straight glossy black hair down to her waist with side-swept bangs, violet eyes, a light lavender sleeveless maxi dress; Boszi: long straight lavender hair, amber eyes, a big navy witch hat with a gold band, a navy strapless dress with small white and gold stars and gold trim. Any phone screen glows with a soft colourful gradient, with no text, no icons and no logos.

Style: a still from a modern 3D animated feature film, stylised characters with big expressive eyes, soft subsurface skin, detailed hair, soft even warm studio light, smooth materials. Behind them a green school chalkboard with no writing on it, and a plain classroom wall.

Check: overlaid at 50% on image 1, there are no double edges on the heads, faces, hair, hands, phones or shoulders.

{ "image_urls": [ "⟨3d-kozep.png⟩", "⟨3d-kozep-szin.png⟩", "⟨eleje.png⟩" ], "image_size": { "width": 1088, "height": 1920 }, "quality": "medium", "num_images": 1, "output_format": "png" }

Stylized 3D animated feature film footage, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-0.9 s: Cépi, the girl with long straight black hair in the lavender dress, stops and looks at her sparkling hand in wonder. 0.9-2.6 s: Cépi pulls out her own phone, raises it high at arm’s length in selfie mode, and starts an enchanted line in awe, her lips move with the voice; she closes her eyes in bliss and puts her other hand to her cheek. Boszi, the girl in the witch hat, quietly leans in behind Cépi’s shoulder to peek into the selfie and smiles; Boszi does not talk. Only Cépi talks. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 2.6, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

Stylized 3D animated feature film footage, one continuous take, 9:16 vertical. The first frame is image 1 and the last frame is the end image: the same people in the same classroom. The camera moves slowly and smoothly in one continuous arc around the group, from the viewpoint of the first frame to the viewpoint of the end frame, at eye level. 0.0-2.5 s: Cépi, the girl with long straight black hair in the lavender dress, holds her phone up in selfie mode and finishes her enchanted line, her lips move with the voice, changing pose once; Boszi in the witch hat peeks into the selfie and fixes her own hair, silent. As the camera moves on, Emóka with the red streak rolls her eyes and folds her arms. 2.9-4.9 s: Kocka with the auburn bun and glasses grins slyly into the lens, says a short mischievous line, her lips move with the voice, and taps her phone screen with one finger. Only one person talks at a time. Everyone keeps moving naturally, nobody freezes. Keep every person's face, hair, clothes, colours and the style exactly as in the first frame. No cuts, no zoom jumps, no soft dissolves or fluid morphs, no extra people, no text appearing.

{ "image_url": "⟨eleje.png⟩", "end_image_url": "⟨vege.png⟩", "target_audio_url": "⟨hang.wav⟩", "duration": 5, "resolution": "768P", "prompt_expansion_mode": "disabled", "seed": 4242 }

A young adult Hungarian woman, about 20 years old, native Hungarian speaker. Vain, dramatic and a little spoiled: a bright, polished, melodic voice like a fairy-tale princess in an animated film, slightly theatrical and posh, very expressive, quick to switch from sweet to offended. Clear studio recording, close microphone, no reverb.

Mi folyik itt? Ezt ti csináljátok? Úristen. Ilyen szép még sosem voltam! Csináld vissza! A szép arcom! Persze. Rajtad eddig se volt mit nézni.

A young adult Hungarian woman, about 20 years old, native Hungarian speaker. Chubby, warm and good-natured, a little clumsy and slow, with a soft, round, friendly voice that easily turns comically whiny and helpless. Clear studio recording, close microphone, no reverb.

Lányok... valaki leengedné a karom? Már egy órája így állok, és senki nem szól semmit. Na jó, akkor majd én megoldom valahogy, mint mindig.