GPT Image 2 prompts

Browse GPT Image 2 prompts for fashion portraits and graphic compositions. Compare how each prompt describes its subject and the way the image is arranged.

探索提示词

9,976 条提示词

Social Media Post - Y2K Street Style Snapshot — output

Social Media Post - Y2K Street Style Snapshot

摆烂程序媛GPT Image 2
A Y2K street corner snapshot featuring two fictional 20-something adult females. Core requirement: The two are standing close at the shoulder and waist, each extending one leg outward with knees bent in a mirror-symmetrical single-leg pose. It’s a natural full-body shot taken by a friend during a shopping or cafe trip, not a commercial. 

Bright noon, summer walk, Y2K revival, natural social media vibe with a mix of poised posing and giggling. No over-refinement, just the atmosphere of a few seconds of collaborative posing. 

Left character: Slim and tall with naturally long legs. Right character: Slightly shorter, naturally built, casual posture leaning towards her friend. 

Expressions: Natural smiles, looking at the camera, squinting slightly from the sun. Outfits: Left wears a white fitted crop T-shirt with a red floral pattern and charcoal Y2K checkered shorts. Right wears an ivory ribbed camisole and light grey loose boxer-style shorts. Accessories include retro sunglasses, pearl necklaces, and Mary Janes/sneakers with loose socks. 

Background: A stone-walled cafe exterior with messy interior details visible through the glass. Lighting: Strong direct noon sunlight with deep shadows and high highlights. Camera: Standard mobile phone wide-angle (26-30mm equivalent) taken from chest height. Vertical 2:3 aspect ratio with light JPEG compression and mobile HDR feel for authenticity.
Profile / Avatar - Summer Portrait with Floral Shadows — output

Profile / Avatar - Summer Portrait with Floral Shadows

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: {argument name="theme" default="Gaze unfolding in flower shadows"} Main Focus: Close-up portrait of a {argument name="person" default="woman in her 20s"} looking at the camera surrounded by small purple, blue, and pink flowers under a blue sky. Large face in center focus, featuring shadows of flower branches on cheeks and neck, wind-blown black hair, and white dress. Person/Expression: Smooth oval small face, wide amber eyes, thin natural eyebrows, clean nose, glossy plump coral pink lips. Head tilted slightly right looking straight at the camera with a quiet dreamy expression and slightly open mouth. Dark brown hair nearly black reaching below collarbone, thin center-parted bangs, fine strands blowing across cheeks and neck. Clothing/Pose: {argument name="outfit" default="soft white thin cotton sleeveless dress"}. Deep V-neck, fine ruffles on straps, small cloth ribbon on left shoulder, fine gathers below chest. Facing front, arms naturally lowered off-screen, standing quietly among flowers. Background/Light: Dense small flowers (purple, blue, pink) and green stems in bottom-left and right foreground, partially blurred. Same flower field in the background with clear blue sky above. Warm afternoon direct sunlight from top-right illuminates face and shoulders, with sharp outlines of thin stems and leaves casting shadows on cheeks, nose, neck, and chest. Golden backlighting on hair edges. Composition/Camera: 3:4 vertical. Chest-up close portrait taken with a medium telephoto lens at eye level. Face placed large in the upper-center, from top of head to chest, with shoulders cut at the sides. Shot through foreground flowers with sharp focus on eyes and lips, with shallow depth of field creating soft round bokeh in both foreground and background flowers. Texture/Style: Photorealistic photo in a summer flower field. High-definition depiction of natural skin texture, faint healthy glow, moist lip gloss, eye reflections, light through individual hairs, fine wrinkles and seams in white cloth, and delicate petals and leaves. Transparent color palette harmonizing warm skin tones, vivid green, and soft purple and blue. Negative: Omission of floral shadows on face; change to background without flowers.
Profile / Avatar - Sun-Dappled Afternoon Profile Portrait — output

Profile / Avatar - Sun-Dappled Afternoon Profile Portrait

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: Faint profile in dappled sunlight

Main: A close-up portrait of a {argument name="subject" default="woman in her 20s"} looking at the camera over her shoulder in warm dappled sunlight. The face is positioned prominently from the center to slightly right, with {argument name="hair color" default="light brown hair"}, {argument name="eye color" default="amber eyes"}, and thin white shoulder straps as the main features.

Person/Expression: A clean oval face, large horizontal amber eyes, soft parallel eyebrows, a thin straight nose, and glossy pale rose lips. The body is turned to the left of the screen while the face is turned toward the camera, making direct eye contact with a calm, transparent expression and a slightly open mouth. Light ash-brown hair down to the collarbone, thin translucent long bangs, tucked behind the left ear, with thin wavy strands falling over the shoulder. Small silver ring and single stud earrings in the left ear.

Clothing/Pose: A white delicate lace sleeveless top. Only the thin shoulder strap is visible on the left shoulder, with the chest area cut off at the bottom. A very thin pale gold chain with a small vertical pendant around the neck. The upper body is angled with the shoulder brought forward, both arms and hands naturally down and out of frame.

Background/Lighting: Vertical wooden pillars and warm-colored buildings in the left background, blue-gray window panes and dark green plants in the far right background, blurred until shapes dissolve. Warm afternoon direct sunlight comes from the upper left, creating bright highlights on the bangs, forehead, bridge of the nose, cheeks, and shoulders. Soft shadows filtered through leaves are left on the left side of the face, with golden backlighting added to the hair outline.

Composition/Camera: 3:4 aspect ratio. A shoulder-to-chest close-up shot with a medium telephoto camera at eye level. Top of the head near the upper edge, face positioned large in the center-right, capturing the left ear and shoulder to the right side of the hair. Hair on the right and chest are cut off by the edges. Sharp focus on both eyes, with background and hair ends significantly blurred due to a very shallow depth of field.

Texture/Style: A photorealistic live-action photograph taken in outdoor natural light. High-definition rendering of fine skin texture and subtle gloss, clear catchlights in the eyes, eyelashes, eyebrows, transmitted light through individual strands of bangs, fresh lips, and small reflections on metal accessories. A clean, natural color tone combining warm skin and hair tones with subtle blue and green in the background.

Negative: Change to short black hair, symmetrical frontal composition.
Social Media Post - Morning Stretch on Tatami Mat — output

Social Media Post - Morning Stretch on Tatami Mat

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: 
{argument name="subject" default="Morning stretch on tatami mats"}

Main Subject: 
A young adult woman sitting on tatami mats in a bright Japanese-style room with morning sunlight coming through the shoji screens, stretching with her legs raised high. The figure is placed large in the center of the screen, focusing on her frontal face, bent legs, soles aligned to the upper left, and hands supporting her legs.

Person/Expression: 
Small oval face, bright smooth skin, large dark brown eyes, thin natural eyebrows, small nose, and pale pink lips. Facing the frontal camera with a friendly, closed-mouth smile. Long eyelashes, thin eyeliner, and distinct peach-colored flush on both cheeks. Dark brown hair, almost black, tied in a messy bun high at the back, leaving thin bangs and thin wisps of hair along the cheeks.

Clothing/Pose: 
{argument name="top" default="Ivory fine rib fabric"} short strap top with small lace edges and a central lace-up ribbon, {argument name="skirt" default="black soft flared mini skirt"}, barefoot. Sitting on the tatami, knees deeply bent and raised high in front of the chest, ankles crossed with soles pointing towards the upper left. Right arm wrapped around the front of the left shin, left hand supporting the leg, upper body visible between the thighs.

Background/Lighting: 
White wooden-framed shoji screens on the left, light brown tatami floor with green borders. A low wooden storage shelf in the back right, a transparent narrow-neck vase with green branches, and a white framed botanical print above. Warm direct morning light enters diagonally from the shoji on the left, softly illuminating the face, shoulders, thighs, and soles, casting soft window frame shadows on the tatami. Diffused light from the white walls fills the shadow side.

Composition/Camera: 
Vertical 3:4. A close-up, near full-body seated portrait captured from a low camera near the tatami floor. The waist and bent legs occupy the lower half, the face is in the upper half to the right, and the crossed soles are in the upper left. The skirt hem is near the bottom edge, and toes are near the top edge. Sharp focus on face, legs, hands, and feet with medium depth of field showing the Japanese room background.

Texture/Style: 
High-definition photorealistic live-action photograph taken in warm morning natural light. Realistic depiction of natural skin texture, stray hairs, ribbed top and lace, black cloth, fingers and soles, tatami weave, shoji paper, wood, and glass vase. A calm color scheme of light wood, white, and black, avoiding excessive skin smoothing or illustrative styles.

Negative: 
Alteration of crossed legs, soles, hand support, facial features, or bun hair.
Social Media Post - Office Printer Break Portrait — output

Social Media Post - Office Printer Break Portrait

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: Taking a break at a multi-function printer

Main Figure: A single {argument name="person" default="young adult woman"} sitting on the glass surface of a large commercial multi-function printer with its lid open in a bright office. Place the person large in the center, featuring the whole body from head to black pumps, {argument name="top" default="pale blue shirt"}, black tie, and {argument name="bottom" default="short pants"}, with the blue-glowing scan surface as the main element.

Character/Expression: A small oval face, bright smooth skin, large dark brown eyes, thin natural eyebrows, small nose, and slightly parted pale coral lips. Faces the camera directly with a quiet and calm expression. Thin eyeliner, long eyelashes, pale pink cheeks. Dark brown straight hair close to black reaching below the chest, thin bangs close to a center part, long hair strands around the face, and hair falling from the right shoulder to the chest.

Clothing/Pose: Very pale blue-grey long-sleeved button shirt with a collar, sleeves folded to below the elbows, and several top buttons open. A thin black tie loosely hanging from the neck, black high-waisted shorts, and glossy black pointed-toe high-heel pumps. Sits on the front edge of the printer's glass surface with both legs spread forward and lowered. Place the left hand on the blue-glowing glass surface and rest the right forearm on the right thigh with fingers hanging naturally.

Background/Light: A large black-and-white multi-function printer from the left to the center, a black document cover opened upwards, and bluish-white light inside. Diagonally placed control panel with a blue round button in the right foreground, and a sheet of paper in the discharge area below. White cabinet, small printer, white-potted indoor plant, and window with blinds in the right background. Strong but soft daylight from the top right window illuminates the face, shirt, and both legs. The blue light from the scan surface reflects faintly on the left hand and lower side.

Composition/Camera: Vertical 3:4. A full-body seated portrait captured from the front from a camera slightly lower than the seat surface. The top of the head is near the top edge, and both pumps are near the left and right bottom edges, placing the person large in the center. Position the trapezoid of the open cover behind and the control panel diagonally at the bottom right. Precisely focused on the face, body, and front of the printer, with a medium depth of field where the background furniture becomes slightly soft.

Texture/Style: High-definition photorealistic live-action photo combining window light and machine light. Realistically depict skin, thin hair strands, thin weave and wrinkles of the shirt, black cloth, gloss of the pumps, resin/glass/LCD/buttons of the printer, and paper. Clean office colors of white, black, and pale blue, avoiding excessive skin smoothing or illustration styles.

Negative: Alteration of facial features, clothing, sitting posture on the printer, or limb positions.
E-commerce Main Image - Lace Corset and Maxi Skirt Wardrobe — output

E-commerce Main Image - Lace Corset and Maxi Skirt Wardrobe

MoodLock|AI画像の惜しいを直すGPT Image 2
Wardrobe: Wardrobe assembly rule: Each FINAL WORN STATE overrides conflicting wearing-state wording only for the item descriptions it explicitly scopes or names, and only for wearing role or use, count, position, side, orientation, routing, fastening, layering, folds, tucking, and draping. Preserve those descriptions for identity, design, material, color, pattern, construction, and proportions. Top: One black-and-warm-ivory lace corset camisole top with a fitted cropped silhouette and structured underwired cups. Two narrow black satin shoulder straps include small silver-tone rectangular sliders and connect separately to the upper cup corners and straight rear edge. Each black satin bust cup is densely gathered along its curved lower seam and trimmed across the upper edge with narrow warm-ivory scalloped floral lace. A small black satin bow centers above a triangular warm-ivory lace bridge between the cups. The torso combines a densely ruched vertical black satin center panel, two long warm-ivory floral-lace side-front panels that taper toward the lower edge, and translucent black mesh side panels. Black satin binding outlines the complete upper and lower edges, while short warm-ivory scalloped lace sections project beneath the outer lower corners. Finish with vertical internal boning channels, a concealed rear hook-and-eye closure, tonal black stitching, and no text, hardware ornaments, or additional graphics. Bottom: One solid black low-rise wrap maxi skirt made from medium-weight cotton twill with a washed matte surface and soft structured drape. A broad asymmetric upper edge closes through two long integral black self-fabric ties anchored at separated front-waist stations; equal usable lengths form one compact center-front knot with two balanced loops and two long tapered tails. The skirt body overlaps into a full A-line silhouette with one broad outer front panel crossing diagonally over a continuous inner panel. Along the lower product-left front, a narrow vertical extension carries four evenly spaced matte-black round buttons and ends in a squared free edge. Near the lower product-right front, add one small silver-white embroidered botanical composition consisting of a compact berry-and-leaf sprig connected to a long fine looping stem that forms one large upper arc and one smaller lower curl. Preserve the unilateral button panel, localized embroidery, clean level maxi hem, reinforced tie anchors, tonal stitching, and absence of additional graphics or hardware. Footwear: One matching mirrored pair of black knee-high pointed-toe boots made from smooth polished leather with slim structured shafts and softly wrinkled ankle transitions. Each boot has a sharply tapered closed toe, a close straight shaft with a subtly lowered rear topline, a fitted heel counter, a narrow high stiletto heel, and a thin black outsole. A concealed full-length zipper follows the inner shaft beneath a narrow leather guard, while fine tonal seams shape the vamp, ankle, and vertical shaft. Preserve identical shaft height, pointed forepart geometry, polished finish, heel profile, and construction across the pair except for anatomical mirroring, with no buckles, laces, graphics, or decorative hardware. Bag: One compact black crescent shoulder bag made from glossy crinkled synthetic leather with a softly structured body, reinforced applied panels, black edge paint, and silver-toned hardware. The broad curved top edge contains one full-width zipper opening with exposed silver teeth, while the lower body expands into a rounded shallow crescent with gathered central volume and a level reinforced base. The visible broad face is organized by a curved top yoke, angular layered corner panels, round metal studs, one short functional buckle tab, and a small shaped flap secured by a round stud; each applied region retains its own stitched perimeter and thickness step. A wide adjustable belt-style strap attaches through metal rings, carries a sequence of large silver eyelets, round studs, and a rectangular buckle, and includes a padded section edged by black braided trim. One slim rectangular auxiliary zip pouch is secured to the lower strap hardware through a short metal ring connector; the pouch has a vertical silver zipper, two round studs, and one raised oval face panel. Preserve the compact crescent proportions, glossy wrinkled face, distinct layered panel geometry, functional zipper paths, and dense coordinated silver hardware. Final worn state: - Bag carry: VIEW: FRONT/NEAR-FRONT SUPPORT IMAGE-RIGHT; BACK/NEAR-BACK SUPPORT IMAGE-LEFT; PROFILE follows wearer-left side. The registered short strap bears weight from that shoulder; bag top sits directly beneath the same armpit against side torso. Both hands stay off; support never moves to forearm, elbow, wrist, or hand. - Finish — Natural: Keep the registered short strap on the wearer-left shoulder and the bag-body top directly beneath the left armpit; adjust only close body contact and bag angle, never the support point. Keep the selected support point exact while allowing limited natural contact variation and one small gravity-led angle. Necklace: A delicate short silver-tone jewelry chain formed as one continuous fine strand with closely spaced micro-links and intermittent tiny elongated bead stations. Keep the chain light, even, and bilaterally balanced from a subtle central station, with no pendant or oversized focal component. Complete it with a compact lobster clasp and a short matching adjustment tail. Maintain uniform link pitch, smooth articulation, secure bead-to-chain joins, restrained polished reflections, and a refined low-profile finish. One medium-length antique-silver pendant necklace constructed as a complete flexible closed circuit. A fine oxidized cable chain with compact oval links closes through a small lobster clasp and short extender. Center one sculptural eye-shaped pendant composed of a narrow pointed oval outer frame, a raised polished upper lid, a dark recessed almond-shaped center, and a tiny silver bead pupil, connected through a compact integrated bail. Preserve the horizontal pendant orientation, dark-and-bright metal contrast, delicate chain scale, and absence of gemstones, lettering, or additional charms. Earrings: One matching pair of large polished warm-gold hoop earrings, shown together as two complete identical products. Each earring forms a slender circular open hoop from uniform round-section metal, with one straight post at the upper terminal and a compact hinged latch at the opposing terminal. Preserve the broad clean circle, lightweight section, mirror-polished warm-gold finish, identical diameter, and absence of gemstones, texture, lettering, or dangling components.
Social Media Post - Summer Festival Selfie Portrait — output

Social Media Post - Summer Festival Selfie Portrait

Prompt アトリエ|AI画像プロンプトGPT Image 2
Subject: Selfie at a sunset festival

Main Figure: A single {argument name="person" default="young adult woman"} taking a selfie with a smartphone extended in the audience area of an outdoor music festival, with a giant stage in the background. Place the person large from the center to the bottom right of the screen, showing the extended right arm, {argument name="top" default="black-grey halter top"}, {argument name="bottom" default="black pleated mini skirt"}, and the audience and stage behind simultaneously.

Character/Expression: A small oval face, bright smooth skin, large dark brown eyes, thin natural eyebrows, small nose, and slightly parted pale coral lips. A calm expression looking at the top right of the screen towards the extended smartphone. Thin black eyeliner, long eyelashes, faint redness at the outer corners of the eyes, and healthy cheek color. Light brown loose wavy hair down to below the shoulders, thin bangs covering the eyes, and hair ends glowing golden in the sunset.

Clothing/Pose: Asymmetric halter neck cropped top with abstract black and off-white floral patterns, multiple thin strings crossing the chest and abdomen, and long tie strings at the sides and hem. Low-rise black pleated mini skirt with golden eyelets and a large black belt. Extend the right arm long towards the smartphone at the top right, lift the left elbow, and place the left hand behind the head. Lean the upper body back slightly and stand with the waist twisted to the left. Blue, yellow, and red festival fabric bands on the right wrist.

Background/Light: A large outdoor stage with black steel trusses in the center background, orange abstract images on the left/right vertical screens and center screen, and many bluish-white lights. In the left/right foreground and middle ground, audiences wearing blackish clothes are densely packed, standing on the grass. The top half is a clear blue sky without clouds. The low sunset at the back left warmly edges the hair and shoulders, with soft front light typical of smartphone photography on the face and body. Shadows are natural, preserving the depth of the audience.

Composition/Camera: Vertical 3:4. A selfie portrait from upper body to thighs captured at a wide-angle from an eye level close to the extended right hand's smartphone. Fit the face in the center, the extended arm and device in the top right, and the skirt in the bottom right, with a wide blue sky overhead and the entire stage at the back left. Precisely focused on the person, with a deep depth of field allowing the audience and stage to be identifiable.

Texture/Style: High-definition photorealistic live-action photo like those taken with a smartphone outdoors in the evening. Realistically depict natural skin texture, wind-blown hair, complex fabric and strings of the top, pleats and metal fittings, fabric bands, audience clothing, steel trusses, and LED screens. Contrast the blue sky and orange stage lights, avoiding excessive skin smoothing or illustration styles.

Negative: Alteration of facial features, clothing strings, selfie arm, or stage placement.
Infographic / Edu Visual - Knolling Teardown Diagram — output

Infographic / Edu Visual - Knolling Teardown Diagram

GadgetifyGPT Image 2
Knolling teardown — {argument name="subject" default="CyberTruck"} ({argument name="subject type" default="EV"}), top-down orthographic. PARTS — complete bill of materials, show each exactly once: {argument name="parts list" default="top housing ×1, bottom housing ×1, main PCB ×1, ribbon cable ×2, lens assembly ×1, M2×5 screw ×6, spring ×2, rubber gasket ×1, ..."} LAYOUT - clean rectilinear grid; every part aligned to horizontal and vertical axes - grouped by subassembly — a subassembly's parts stay adjacent as one cluster - each group anchored by its largest structural part; fasteners cluster beside the part they secure - even spacing, no overlapping parts, no diagonal scatter - consistent scale across every part DEPICTION — keep it truthful - every fastener sits near a matching mounting hole; screw count = hole count - every gasket or seal shown with the perimeter it seals against - no duplicated or phantom parts - preserve each part's true orientation and handedness; do not mirror chiral parts RENDER - orthographic projection, no perspective, all parts in sharp focus - flat, even, diffuse lighting; minimal soft shadow - neutral seamless background
Infographic / Edu Visual - Blueprint to Model Desk Diorama — output

Infographic / Edu Visual - Blueprint to Model Desk Diorama

GadgetifyGPT Image 2
Function DrawArchitecture(input building/architect)
Input Variable: {argument name="building or architect" default="Fallingwater"}

Generate a hyper-realistic, 1:1 product photo of a "Blueprint-to-Model Desk Diorama" (Isometric miniature).

1. Semantic Extraction:
- Identify style (brutalist, modernist, gothic), key forms, and one signature feature (cantilever, arch, atrium).
- Figure: Chibi “architect” avatar holding a tiny scale ruler.
- Finish: Matte PVC.

2. Container (Drafting Corner Box):
- Structure: U-shaped cardboard drafting station (two walls + floor), desktop scale.
- Props: tiny triangle ruler, pencil, scale bar.

3. Background (Blueprint Walls):
- Walls: B&W blueprint lines, elevation drawings, annotations (original, inspired).
- Floor: Blueprint fragments “explode” upward into paper folds.

4. Integration:
- A 2D blueprint line rises into a 3D model piece (a wall or arch popping out) that the figure points at.

5. Visual Syntax:
- Palette: Strict B&W blueprint world; full color figure + one accent prop.
- Lighting: soft product photo, crisp details.
- Label: "[Subject] — [Style/Architect]"

Return: ONE image, 1:1, museum-miniature realism.

<instructions>
Input: building/architect
Analyze style, era, signature geometry. Find 4 MUCH lesser-known buildings/architects in same lineage.
Output: a 2x2 grid, each panel is DrawArchitecture for a lesser-known recommendation.
</instructions>
Profile / Avatar - High-Angle Editorial Editorial Portrait — output

Profile / Avatar - High-Angle Editorial Editorial Portrait

EeshaGPT Image 2
A clearly adult woman in a high-angle close-up, framed from forehead to collarbone. Long dark center-parted hair softly frames her face. {argument name="eye color" default="Light green eyes"} are the main focus, with natural lashes, brows and catchlights. 

Realistic light skin shows pores, tiny moles, subtle asymmetry and a soft pink cheek flush. Muted pink lips are relaxed and slightly parted. Soft overhead lighting creates gentle highlights on the forehead, nose and cheekbones with natural shadows.

Pale light-blue background with a crisp white collar entering from the left. Ultra-photorealistic beauty portrait, 85mm, f/1.4, shallow depth of field, natural skin texture, refined editorial clarity. --ar 4:5 --quality high --no altered eye color, mismatched eyes, extra eyes, distorted face, plastic skin, excessive retouching, artificial symmetry, harsh HDR, CGI, cartoon, heavy blur.
Social Media Post - Cross-armed toast interaction prompt — output

Social Media Post - Cross-armed toast interaction prompt

Curline ZephirinGPT Image 2
Draw a scene of {argument name="people" default="Lei Jun and Dong Mingzhu"} having a {argument name="cross-armed toast" default="cross-armed toast"} on the {argument name="location" default="Eiffel Tower"}
YouTube Thumbnail - Japanese ASMR Streamer Studio — output

YouTube Thumbnail - Japanese ASMR Streamer Studio

Cheer CheungGPT Image 2
A {argument name="person" default="Japanese female streamer"} speaking into an {argument name="microphone" default="ASMR microphone"} in a live streaming studio.
Profile / Avatar - Shy Red-Haired Anime Realism Portrait — output

Profile / Avatar - Shy Red-Haired Anime Realism Portrait

KISA aka Copenzafan.ethGPT Image 2
Create an intimate close-up portrait of {argument name="character name" default="a young woman"} with a stylized semi-realistic CGI/anime-realism look blended with photographic skin detail. She has {argument name="hair color" default="deep crimson red"} shoulder-length messy hair swept across her forehead, loose strands framing her face, pale skin, large glossy hazel-green eyes, long lashes, dark winged eyeliner, softly arched worried eyebrows, a small nose, and slightly parted natural pink lips. Her expression is shy, anxious, and tender, looking slightly off to the viewer’s right rather than directly at the camera. Add visible anime-style blush marks: exactly 18 diagonal pink strokes across both cheeks and the bridge of the nose, with 8 strokes on the viewer-left cheek area, 2 shorter strokes over the nose bridge, and 8 strokes on the viewer-right cheek area. Frame her from bare shoulders upward in a tight vertical portrait, face filling most of the canvas, with the top of the hair close to the upper edge and shoulders softly visible at the bottom. Use shallow depth of field, soft moody indoor lighting from the upper left, subtle catchlights in the eyes, realistic pores and skin texture, gentle subsurface scattering, cinematic contrast, muted dark gray-brown background blur, and a high-end character-render-meets-photography finish. Keep the image clean with no text, no accessories, and no extra characters.
Comic / Storyboard - Resolution: — output

Comic / Storyboard - Resolution:

HemanthGPT Image 2
Labeled production character design sheet for MASKED COURIER, an original young adult rooftop runner in a Neo-Edo poster illustration style. Create one clean landscape
Infographic / Edu Visual - LLMFIT Model Loadout Dashboard — output

Infographic / Edu Visual - LLMFIT Model Loadout Dashboard

Rich · Atom Tan StudioGPT Image 2
Goal: Create a dark futuristic dashboard infographic for {argument name="dashboard title" default="LLMFIT RECOMMENDATIONS"}, showing local AI model recommendations for a workstation.

Canvas: Wide 21:9 desktop-panel image, black and deep teal background with subtle glow, thin neon cyan border, faint scanline/grid texture, compact technical UI styling.

Layout: Top header bar with small label “LEGION / MODEL INTELLIGENCE”, large title “LLMFIT RECOMMENDATIONS”, subtitle “NVIDIA GeForce RTX 5090 · 31.84 GB VRAM · 125.18 GB RAM”, and a small outlined close-box icon at the top right. Main content is split into two panels: a large left recommendation module occupying about two thirds of the width, and a narrower right verification panel.

Left panel: Add a large cyberpunk card titled “LEGION MODEL LOADOUT”, with “LEGION” in neon lime and the rest in white blocky techno typography. Under the title, show exactly 4 hardware/status tiles with icons: 1) NVIDIA GeForce RTX 5090, 31.8 GB VRAM with a GPU fan icon, 2) Intel(R) Core(TM) Ultra 9 285K with a CPU chip icon, 3) 125.2 GB system RAM with a memory module icon, 4) CUDA with a circular CUDA emblem. Beneath the tiles, show exactly 6 ranked model rows, each with a large lime outlined rank number, model name, quantization, runtime, RAM amount, and estimated tok/s speed. The 6 rows are: 1) shawnw3j/Huihui-Qwen3.6-27B-abliterated-AWQ-MTP — AWQ-4bit — vLLM — 14.7 GB — 80.9 estimated tok/s; 2) Vortex5/G4-Starry-Ocean-12B — Q8_0 — llama.cpp — 16 GB — 82.8 estimated tok/s; 3) shawnw3j/Qwen3.6-27B-AWQ-MTP — AWQ-4bit — vLLM — 14.7 GB — 80.9 estimated tok/s; 4) Minachist/Qwen3.6-27B-INT8-Autoround-V2 — AutoRound-4bit — vLLM — 16.6 GB — 80.9 estimated tok/s; 5) exnivo/Qwen3.8-20B-Minitron — Q8_0 — llama.cpp — 22.6 GB — 49.9 estimated tok/s; 6) Lorbus/Qwen3.6-27B-int4-AutoRound — AutoRound-4bit — vLLM — 16.6 GB — 80.9 estimated tok/s. Use small cyan icons for chip/runtime/RAM/speed columns.

Footer inside left card: Center a slim neon divider with the text “ESTIMATED BY LLMFIT · VERIFY WITH A LOCAL BENCHMARK.”

Right panel: Title it “VERIFIED LLMFIT DATA” with a small note “ESTIMATES, NOT BENCHMARKS”. Show exactly 6 compact verification entries matching the same 6 models, numbered 01 through 06 in lime, each with smaller gray metadata text and a bright cyan score on the far right: 80.9, 82.8, 80.9, 80.9, 49.9, 80.9. Add a small orange warning note at the bottom: “llmfit recommendations are estimates from detected hardware, not measured benchmarks.”

Bottom app chrome: Add a tiny timestamp line at bottom left, “GENERATED 8/17/2026, 7:53:32 PM”, and a small green outlined button at bottom right labeled {argument name="button label" default="Refresh scan"} with a refresh icon.

Visual style: High-contrast sci-fi terminal UI, angular panel corners, thin glowing cyan circuit traces, lime accents, white condensed techno font, dense but readable technical typography, subtle green monitor glow. Keep the image crisp like a generated dashboard screenshot, not a poster.

Constraints: Use exactly 6 model recommendation rows, exactly 4 hardware tiles, and exactly 6 verified-data entries. Do not add people, photos, logos beyond simple hardware-style icons, or extra sections. Keep all visible text in English.
Profile / Avatar - Cyberpunk Neon Ninja Samurai — output

Profile / Avatar - Cyberpunk Neon Ninja Samurai

MarkGPT Image 2
Create a vertical cinematic cyberpunk action portrait of a lone futuristic ninja-samurai standing in a rain-soaked neon city street at night. The subject is a full-body male figure in a black hooded tactical outfit with layered leather armor, straps, utility harnesses, armored forearms, gloves, cargo pants, and heavy combat boots. His face is hidden by a glossy black cyber mask with glowing neon circuitry and two bright X-shaped eyes, one cyan and one magenta; the hood casts deep shadow over the mask. He holds exactly one katana in his right hand, angled diagonally downward across the foreground, with a glowing energy blade wrapped in orange fire and blue electrical arcs. A second sheathed sword is visible at his left hip/back. Surround him with dramatic swirling elemental energy: one large orange flame ribbon looping from the left side behind his shoulder down around the sword, and one bright cyan lightning stream crossing behind and around him, with many small sparks and embers. The environment is a dark futuristic alley or street with blurred blue and orange neon signs, wet reflective pavement, rain mist, and bokeh city lights in the background. Use high contrast teal-and-orange color grading, volumetric haze, rim lighting, realistic glossy reflections, cinematic depth of field, sharp detailed character textures, and dynamic motion in the energy effects. The character should feel intimidating, calm, and powerful, standing upright and facing slightly to the right. Customize the main figure as {argument name="character name" default="cyberpunk hooded ninja samurai"}, the outfit color as {argument name="outfit color" default="black"}, the mask eye colors as {argument name="mask eye colors" default="cyan and magenta"}, the sword energy colors as {argument name="sword energy colors" default="orange fire and blue lightning"}, and the setting as {argument name="setting" default="rainy neon cyberpunk city street"}. No text, no watermark, no extra characters, no cartoon style.
Infographic / Edu Visual - Industrial Heritage Open Day Poster — output

Infographic / Edu Visual - Industrial Heritage Open Day Poster

小小东GPT Image 2
Goal: Create a vertical industrial heritage opening-day poster for {argument name="event name" default="OPEN DAY"}, combining oversized brutal typography, Chinese museum-poster text, architectural cross-section drawing, ink stains, and aged printed-paper texture.

Canvas: Portrait 3:4 poster, off-white warm paper background with visible grain, scratches, speckles, faint fold lines, light gray smudges, and subtle letterpress/risograph misregistration. Use a restrained palette of black, charcoal gray, faded beige, rust brown, and thin red-brown annotation lines.

Layout: Dominant typography fills the upper and left side: huge cropped black condensed sans-serif letters reading “OPEN” across the top and “DAY” beneath it, so large that the words are partially cut off by the canvas edges and overlap the architecture. Center the composition on a detailed old factory building elevation/cutaway, occupying most of the lower two-thirds. Place a tall chimney on the right side of the building, rising into the upper area. Keep the right margin for vertical Chinese title text and small annotations. Keep the left margin for a numbered guide list and opening-hour details.

Main subject: A ruined industrial site rendered as a precise architectural ink-and-watercolor section drawing. Show a tall multi-level factory frame with exposed steel trusses, scaffolding, platforms, staircases, old brick and concrete walls, large rectangular windows, arched ground-floor openings, pipes, tanks, small machinery, rail-cart-like objects, and tiny human figures for scale. Add a very tall cylindrical striped smokestack/chimney to the right, weathered and gray. The building should feel like an open industrial relic being preserved and regenerated, with transparent cutaway floors and visible internal structure.

Text content: Include exactly 6 numbered guide entries on the left, each with a large two-digit number and bilingual-style small labels: 01 Main Entrance, 02 Boiler House, 03 Workshop, 04 Machine Shop, 05 Chimney / Water Tower, 06 Viewing Platform. Include small time ranges next to entries, such as 09:30 / 13:30, 10:00 / 14:00, 10:40 / 14:40, 11:20 / 15:20, 12:00 / 16:00, 12:30 / 16:30. Add an opening-hours block reading “OPENING HOURS 09:30 — 17:00” with smaller Chinese text beneath. Add an address block with small Chinese text. On the right side, add large vertical Chinese characters for {argument name="Chinese title" default="开工 开放日 工业遗址"}, arranged in three stacked columns. Add a small caption reading “过去 / 现场 / 未来” and a short Chinese slogan below it. Add tiny English words near the upper right: “SITE / MEMORY / FUTURE”. Near the lower right add tiny stacked words: “PRESERVE / REGENERATE / SHARE”.

Annotations: Include exactly 3 thin red-brown circular callout graphics labeled 02, 03, and 05 positioned around the building; also include one separate circular date mark near the upper right reading {argument name="date mark" default="05 / 18"}. Add delicate leader lines, small tick marks, microscopic labels, and a few year notes such as 1958, 2012, and 2024.

Visual style: Industrial design poster, architectural blueprint meets ink-wash illustration, archival museum print, worn offset printing, black distressed typography, gridded but asymmetrical Swiss poster composition, subtle rust stains and ink bleeds. The typography should look bold, printed, slightly rough, and physically integrated with the paper. Keep all elements flat and poster-like, not photorealistic.

Constraints: Use exactly one main factory building, exactly one tall chimney, exactly 6 guide-list items, exactly 3 numbered architectural callout circles plus 1 date circle. Avoid bright colors, glossy effects, modern glass architecture, extra logos, QR codes, or watermarks.
Product Marketing - Luxury SPA Service Menu Poster — output

Product Marketing - Luxury SPA Service Menu Poster

小小东GPT Image 2
Generate a luxury beauty promotion poster centered on user requirements. A pure white rounded rectangular price card with rich 3D layers floats in the lower-middle section, featuring soft diffused shadows and overlapping back layers, with a circular punched tag detail in the upper left corner. The top half of the poster is a transparent and ethereal light pink soft-light textured background, with subtle fluid marble and petal-blending textures, complemented by a faint light-colored postmark watermark on the right edge. A high-contrast modern serif uppercase English title is centered at the top, followed by a Chinese slogan decorated with curly braces and ultra-fine subtext. The floating card presents a rigorous and clear information architecture: the top segment has the theme card name and fine hand-drawn embellishments; below is a high-contrast focal price area with currency labels on the left and oversized modern bold serif italic numbers in bright rose pink on the right. The middle segment contains champagne-pink gradient rounded capsule feature tags, and the lower segment neatly lists multiple equidistant project items, each guided by tiny dots, including project name and frequency, aligned on the right with uniform black rounded capsule unit price tags with white text, separated by ultra-fine light grey lines. The card bottom ends with a smooth, elegant light pink upward-convex arc, with a micro-convex 3D embossed heart badge and small uppercase English text in the center. The bottom edge of the poster leaves comfortable light pink whitespace with a centered, widely spaced micro-slogan text. The overall color scheme is based on soft light pink, pure white, and high-contrast black capsules, with pure high-key light and precise lines, showcasing an elegant and romantic commercial visual.

Theme: {argument name="theme" default="SPA Club Service Price List"}
Forget the original colors, re-plan a color scheme suitable for the theme. I like {argument name="color palette" default="purple"}
Ratio 9:16
Profile / Avatar - Stern Hamster Stop Gesture — output

Profile / Avatar - Stern Hamster Stop Gesture

🐹🐹GPT Image 2
Create a square digital illustration of a chubby golden Syrian hamster doing the “talk to the hand / stop” gesture, with one small pink paw raised upright beside its face like a human palm. The hamster has {argument name="fur color" default="warm golden orange and creamy white"} fur, a rounded oversized body, fluffy cheek and belly fur, small rounded ears with darker inner shading, glossy narrow black eyes angled in a stern unimpressed expression, a tiny pink nose, fine white whiskers, and a slightly downturned mouth. Pose it in a three-quarter front view, leaning slightly forward, with the raised paw on the viewer’s right and the other paw hanging low. Use detailed soft anime-style rendering with visible fur strokes, gentle painterly shading, clean dark linework, and a cute-but-serious comedic mood. Place the character large in frame, cropped close so the body fills most of the square canvas, against a simple flat {argument name="background color" default="bright warm yellow"} background with no props, no text, and no watermark.
Comic / Storyboard - Neon Anime Stairway Encounter — output

Comic / Storyboard - Neon Anime Stairway Encounter

𝟡𝟜 ᴾᴸᴬʸᶠᴼᴿᴳᴱGPT Image 2
Create a dramatic vertical anime-style illustration of two stylish young adults on an outdoor apartment stairway, seen from a low tilted angle with strong perspective and cinematic depth. The scene contains exactly 2 characters: 1 blonde young woman in the foreground and 1 dark-haired young man higher up the stairs. The woman has long messy flowing {argument name="hair color" default="pale blonde hair"}, a soft melancholic gaze looking back toward the viewer, a loose oversized white shirt, wide black trousers, and a dark jacket hanging from one arm; her body is turned away while her face turns over her shoulder. The man has tousled black hair, glasses, a serious distant expression, a black button-up shirt with sleeves rolled up, and slim black trousers; he stands one or two steps above her, looking down toward her. Place them on a red-and-blue lit exterior staircase with exactly 4 visible stair levels, a metal railing, a bright white sky with soft cyan clouds, 1 open red-edged door on the upper right, and 1 small shoe shelf beside the door holding several shoes. Use intense stylized lighting: saturated cyan daylight from the sky, hot magenta-red rim light and shadows across the stairs, high contrast cel shading, neon color separation, sharp ink-like linework, glossy highlights on hair and fabric, and a slightly blurred lower foreground foot/step for motion and depth. Composition should feel like a fashionable romantic tension scene, wide-angle lens distortion, diagonal architecture, elegant anime concept art, clean white negative space in the sky, no added captions or watermark.
Comic / Storyboard - Anime Imperial Sniper Commander — output

Comic / Storyboard - Anime Imperial Sniper Commander

癒音ちー✨ゆおんちー✨癒やし声ASMRとAIGPT Image 2
Create a dramatic vertical anime-style illustration of a smug young female military commander standing in a ruined battlefield, viewed from a low angle to make her look dominant and intimidating. She has a short light-brown bob haircut with straight bangs and vivid pink eyes, wearing a confident half-smile. Her outfit is an ornate black imperial officer dress uniform with glossy fabric, gold embroidery, gold buttons, braided cords, medals on the chest, a white belt with a gold buckle, black gloves, epaulettes with gold fringe, a dark red cape lining, garter straps, and thigh-high black stockings. She holds an enormous black anti-materiel sniper rifle with a long barrel and large scope slung diagonally across her shoulders, the weapon dominating the upper left-to-right composition. The background is a devastated gothic war zone with broken stone ruins, jagged rubble in the foreground, torn crimson banners, distant spires, smoke, embers, and a blazing backlit cloudy sky with sunbeams breaking through. Use high-detail cinematic lighting, strong rim light, warm fire glow, dark shadows, polished metal highlights, dynamic cape movement, floating sparks, ultra-detailed anime rendering, sharp linework, rich contrast, and a powerful revenge-themed mood. No text, no watermark, no extra characters.
Social Media Post - Two Reference Women at Rural Train Station — output

Social Media Post - Two Reference Women at Rural Train Station

千田結穂(せんだゆいほ)GPT Image 2
Using REFERENCE_1 as the woman on the left and REFERENCE_0 as the woman on the right, generate a realistic full-body travel snapshot of the two distinct people standing together at a quiet Japanese countryside train station platform. Keep each face and hairstyle clearly separated and faithful to its own reference; do not blend their facial features.

Scene and composition: Portrait-oriented photo, both women standing side by side near a dark wooden station pillar, looking at the camera with soft natural smiles. Add railway tracks running into the distance on the left, a rural platform, station buildings, green hills in the background, and bright daytime natural light.

Wardrobe and props: Change both outfits from plain gray T-shirts into soft, modest travel clothing. The left woman wears a white long-sleeve blouse and pale blue skirt, with a beige crossbody strap, and holds the handle of a light-colored rolling suitcase in front of her. The right woman wears a white top with a pale sage-green cardigan and cream trousers, carrying a light beige shoulder bag on her right shoulder.

Style: Natural photorealistic Japanese lifestyle photography, gentle colors, shallow depth of field, clean realistic skin texture, candid but polished magazine-like travel atmosphere.

Constraints: Exactly 2 women, matching REFERENCE_1 on the left and REFERENCE_0 on the right; no extra people, no text, no face swapping, no mixed identities.
Comic / Storyboard - Anime Tactical Girl in Sunlit Bunker — output

Comic / Storyboard - Anime Tactical Girl in Sunlit Bunker

癒音ちー✨ゆおんちー✨癒やし声ASMRとAIGPT Image 2
Create a polished anime-style vertical portrait of a young female tactical officer in a ruined concrete military shelter, shown from mid-thigh up and turned three-quarters away while looking back over her shoulder with a gentle, confident smile. She has {argument name="hair color" default="light ash-brown"} short bobbed hair with soft bangs and flyaway strands, large {argument name="eye color" default="pink"} eyes, fair skin, and a subtle blush. Her outfit is an elegant black military coat with gold trim, brass buttons, a high collar, decorative shoulder cords, an eagle-like gold insignia on the sleeve, a fitted tactical belt with pouches, black gloves, a short skirt edge visible under the coat, thigh straps, and black thigh-high stockings. She holds one modern black assault rifle diagonally across her body with both hands, detailed with a scope, rail system, magazine, suppressor-like barrel attachment, and realistic mechanical parts. The scene is set in {argument name="setting" default="a war-torn concrete bunker with sandbags and stacked equipment crates"}, with cracked walls, a dark ceiling, dusty air, and rubble in the background. Warm sunlight streams dramatically from the upper right in visible beams, creating glowing dust particles, rim light on her hair and coat, soft lens bloom, and high-contrast cinematic shadows. Use highly detailed Japanese anime illustration quality, delicate linework, semi-realistic lighting, rich fabric folds, metallic weapon details, dynamic coat movement sweeping left, shallow depth of field, and a warm sepia-gray color palette. No text, no watermark, no extra characters, exactly one girl and exactly one rifle.
Profile / Avatar - Angular Manga Character Redesign — output

Profile / Avatar - Angular Manga Character Redesign

抹茶オレンジ🧋🍊GPT Image 2
Using REFERENCE_0 as the character base, create a clean standalone full-body manga illustration instead of the character-design sheet. Remove all annotation text, guide diagrams, inset panels, sample expressions, tone swatches, and background comic panels. Regenerate the same boy in a more angular, gambling-manga-inspired black-and-white style with sharper facial geometry, narrow triangular eyes, a pointed nose, harsher zigzag contour lines, and heavier dramatic ink shadows. Pose him standing front-facing with a slight contrapposto lean, head tilted downward, both hands in his pants pockets, giving a cool aloof expression. Update the outfit to an open light jacket with a high popped collar over a black shirt, dark straight pants with a belt, and oversized chunky shoes. Keep the spiky hair silhouette and monochrome manga screen-tone rendering, but make the final image a single centered character on a plain white background with no text, no panels, and no extra objects. Character style target: {argument name="manga style influence" default="Kaiji-inspired angular seinen manga"}. Outfit color balance: {argument name="outfit contrast" default="light open jacket, black shirt, dark pants"}. Pose: {argument name="standing pose" default="front-facing, head slightly lowered, both hands in pockets"}. Background: {argument name="background" default="plain white"}.
按模型、风格或主题浏览

寻找下一个方向

ASTORIE CANVAS

让灵感,继续生长。

在同一个创作空间,连接提示词、参考素材和生成结果。

打开创作画布 ↗