GPT Image 2 Prompts Gallery

image_settings.aspect_ratio: 3:4 image_settings.resolution.width: 1152 image_settings.resolution.height: 1536 prompt.identity_lock.reference: input_photo prompt.identity_lock.preserve[1]: face prompt.identity_lock.preserve[2]: facial proportions prompt.identity_lock.preserve[3]: eye shape prompt.identity_lock.preserve[4]: nose prompt.identity_lock.preserve[5]: lips prompt.identity_lock.preserve[6]: skin tone prompt.identity_lock.preserve[7]: skin texture prompt.identity_lock.preserve[8]: hairline prompt.identity_lock.preserve[9]: overall identity prompt.identity_lock.rules[1]: no face swap prompt.identity_lock.rules[2]: no beautify prompt.identity_lock.rules[3]: no smoothing prompt.identity_lock.rules[4]: no reshaping prompt.identity_lock.rules[5]: no AI look prompt.scene.camera_angle: high-angle downward shot, POV prompt.scene.composition: MacBook screen fills most of the frame, thin strip of physical keyboard visible at the bottom prompt.scene.screen_surface[1]: visible RGB pixel grid prompt.scene.screen_surface[2]: subtle moire effect prompt.scene.screen_surface[3]: micro dust on glass prompt.scene.screen_surface[4]: faint fingerprints prompt.scene.screen_surface[5]: soft ambient reflections prompt.digital_interface.os: macOS dark mode prompt.digital_interface.background_app.name: Spotify prompt.digital_interface.background_app.view: Liked Songs prompt.digital_interface.background_app.visible_tracks[1]: Blank Space – Taylor Swift prompt.digital_interface.background_app.visible_tracks[2]: Shake It Off – Taylor Swift prompt.digital_interface.background_app.visible_tracks[3]: Cruel Summer – Taylor Swift prompt.digital_interface.background_app.visible_tracks[4]: Love Story – Taylor Swift prompt.digital_interface.foreground_app.name: Photo Booth prompt.digital_interface.foreground_app.state: live preview window prompt.digital_interface.foreground_app.position: floating, center-right prompt.photo_booth_content.environment.room: dim bedroom prompt.photo_booth_content.environment.background: off-white wall, rumpled bedding prompt.photo_booth_content.environment.lighting: low-light, nocturnal, cool screen glow mixed with warm skin tones prompt.photo_booth_content.subject.pose: lying down, relaxed, candid prompt.photo_booth_content.subject.expression: natural, slightly relaxed prompt.photo_booth_content.subject.outfit: light-colored tank top prompt.photo_booth_content.subject.prop.item: iPhone 15 Pro prompt.photo_booth_content.subject.prop.hand: right hand prompt.realism_rules[1]: this is a photo of a screen, not a screenshot prompt.realism_rules[2]: raw smartphone photo look prompt.realism_rules[3]: natural noise prompt.realism_rules[4]: imperfect glass prompt.realism_rules[5]: no studio lighting prompt.realism_rules[6]: no HD polish negative_prompt[1]: screenshot negative_prompt[2]: flat UI negative_prompt[3]: perfect screen negative_prompt[4]: clean glass negative_prompt[5]: studio lighting negative_prompt[6]: beauty filter negative_prompt[7]: cartoon negative_prompt[8]: 3d render negative_prompt[9]: painting negative_prompt[10]: watermark negative_prompt[11]: blurred face

Youtube screenshot of an openai livestream where sam altman and rahul gandhi announce that rahul is the new CTO

Subject & Composition: Central Figure: A sharp profile view of a man wearing a classic 1940s fedora and a formal suit. Melting Effect: The lower half of the man's torso should dissolve or "drip" into long, vertical black ink streaks and splatters, creating a surreal, liquid-bleed effect against the background. Graphic Element: A large, solid vibrant red circle is positioned directly behind the man's head and shoulders, creating a "halo" or "rising sun" motif that anchors the composition. Style & Color Palette: Color Scheme: A strict minimalist palette of True Red, Deep Black, and Off-White/Cream. Photography Style: High-contrast, black and white film noir photography for the subject. The texture should look like a high-resolution fashion editorial with a shallow depth of field. Background: A clean, slightly textured off-white paper or canvas background to give it a premium, physical poster feel. Technical Details: Lighting: Strong "Rembrandt" lighting to define the facial contours, emphasizing the bridge of the nose and the jawline with deep shadows. Atmosphere: Somber, intellectual, and high-stakes. The design should feel modern yet period-accurate to the mid-20th century. Framing: Vertical poster orientation (A-frame), with significant negative space on the left and right to emphasize the central vertical axis. Aspect ratio is 9:16

[STYLE] Monochrome grayscale illustration, 3D-rendered character, clean instructional reference sheet, white background, comic-style cell grid layout, technical diagram aesthetic. [LAYOUT] 4×4 grid layout with a total of 16 panels. Each panel is separated by thin black border lines. Cells are numbered from 1 to 16, with consistent panel sizes. [CHARACTER] image1 (the same character appears consistently in all panels) [PANEL STRUCTURE – per cell] Top-left: bold number badge + English title text Center: full-body character pose illustration Bottom-left: English description text (3–4 lines) Overlay: directional arrows indicating movement [ARROWS / MOTION INDICATORS] Curved arrows, straight arrows, and circular rotation indicators placed around the character to show motion flow and direction. [RENDERING STYLE] Highly detailed 3D sculpted style, soft studio lighting, subtle shadows, no color, grayscale shading, clean linework, game concept art quality. [NEGATIVE] No background scenery, no color tones, no additional characters, no complex background.

subject.description: Young woman with a fit yet voluptuous physique, appearing in her 20s. She has a fair complexion with visible freckles across her nose and cheeks. Her face is framed by messy, light brown hair styled in an updo with long, wispy bangs covering her forehead. subject.body_details.bust: Extremely voluminous and heavy bust, clearly significantly above average size. The breasts appear full with a natural, heavy teardrop shape, creating deep cleavage. The volume projects forward and outward, dominating the lower center frame. The skin on the chest is sun-kissed with visible texture. subject.body_details.torso: Visible collarbones and neck, leading down to the large bust. The waist is obscured by the close framing. subject.body_details.skin: Tanned skin tone with natural sun damage, freckles, and skin texture. Not airbrushed. subject.clothing.top: White, scoop-neck sports bra or crop top made of a smooth, matte fabric. The straps are thin. The fabric is stretched tight across the chest volume. subject.clothing.accessories.necklace: A simple gold necklace with a circular pendant resting on the upper chest. Subtle gold bracelet on the left wrist. Small stud earrings. pose.type: Seated selfie portrait pose.head: Facing directly forward, chin slightly tucked, gaze directed straight into the camera lens with a pouty expression. pose.arms_and_hands: Her right arm (viewer's left) extends forward out of frame, indicating she is holding the camera. Her left hand (viewer's right) is raised to her shoulder, with fingers (featuring a French manicure) hooking under the bra strap and pulling it slightly outward and upward. pose.posture: Leaning slightly forward towards the camera, emphasizing the depth of the chest relative to the face. environment.location: Interior of a luxury vehicle. environment.details: Red leather seats with visible stitching behind her. Through the rear window, a sunny outdoor scene is visible, including a paved road, palm trees, and the rear of a white sedan. environment.context: Daytime travel setting. camera.perspective: Close-range selfie perspective (arm extension implied). High angle looking slightly down at the face and torso. camera.focal_length: Wide angle (approx 24mm to 35mm), creating a slight barrel distortion that accentuates the foreground (chest) and facial features. camera.depth_of_field: Shallow to medium depth of field; the subject is sharp, the background through the car window is slightly soft but distinguishable. lighting.source: Natural daylight entering through car windows. lighting.quality: Bright, somewhat harsh sunlight creating high contrast. Strong highlights on the forehead, nose, and upper curves of the breasts. lighting.shadows: distinct shadows cast by the bangs onto the forehead and eyes. Shadows defining the cleavage and the contours of the collarbones. mood_and_expression.emotion: Confident, sultry, slightly bored or posing casually. mood_and_expression.facial_features.lips: Plump lips in a pout mood_and_expression.facial_features.eyes: blue-grey eyes looking intensely at the viewer mood_and_expression.facial_features.eyebrows: relaxed eyebrows style_and_realism.genre: Casual selfie, social media realism. style_and_realism.rendering: Photorealistic, high fidelity to texture and imperfections. style_and_realism.aesthetic: Raw, unfiltered day-in-the-life snapshot. colors_and_tone.palette: Warm skin tones, bright white top, vibrant red leather seat, muted greens and blues in the background. colors_and_tone.contrast: High contrast due to direct sunlight. colors_and_tone.saturation: Natural saturation, slightly warm color temperature. quality_and_technical_details.resolution: High quality_and_technical_details.sharpness: Sharp focus on eyes and chest, texture visible. quality_and_technical_details.noise: Low ISO, clear daylight shot. aspect_ratio_and_output.ratio: 3:4 aspect_ratio_and_output.framing: Vertical portrait crop, cutting off at the top of the head and just below the bust line. controlnet.pose_control.model_type: DWPose controlnet.pose_control.purpose: Exact skeletal and pose lock controlnet.pose_control.constraints[1]: preserve hand pulling strap gesture controlnet.pose_control.constraints[2]: preserve head tilt controlnet.pose_control.constraints[3]: preserve shoulder alignment controlnet.pose_control.recommended_weight: 1.0 controlnet.depth_control.model_type: MiDaS controlnet.depth_control.purpose: Depth, volume, and camera-to-body spatial lock controlnet.depth_control.constraints[1]: preserve chest projection relative to the face controlnet.depth_control.constraints[2]: maintain car interior depth planes controlnet.depth_control.recommended_weight: 0.8 negative_prompt.forbidden_elements[1]: anatomy normalization negative_prompt.forbidden_elements[2]: body proportion averaging negative_prompt.forbidden_elements[3]: smaller bust than reference negative_prompt.forbidden_elements[4]: reduced chest volume negative_prompt.forbidden_elements[5]: flattened breasts negative_prompt.forbidden_elements[6]: compressed chest negative_prompt.forbidden_elements[7]: artificial lift negative_prompt.forbidden_elements[8]: slimmed torso negative_prompt.forbidden_elements[9]: aesthetic proportion correction negative_prompt.forbidden_elements[10]: beauty standard enforcement negative_prompt.forbidden_elements[11]: dataset-average female anatomy negative_prompt.forbidden_elements[12]: mirror selfie negative_prompt.forbidden_elements[13]: phone visible in hand negative_prompt.forbidden_elements[14]: reflections of phone negative_prompt.forbidden_elements[15]: beautification filters negative_prompt.forbidden_elements[16]: skin smoothing negative_prompt.forbidden_elements[17]: plastic skin negative_prompt.forbidden_elements[18]: airbrushed texture negative_prompt.forbidden_elements[19]: stylized realism negative_prompt.forbidden_elements[20]: editorial fashion proportions negative_prompt.forbidden_elements[21]: studio lighting

Assemble the dieline into a flawless 3D box with accurate panels, clean folds, undistorted type, and artwork preserved exactly. shoot it upright at a refined ¾ angle in a minimal premium studio setting with a soft neutral background, diffused light, subtle shadows, no props, true colours, matte paperboard texture, and realistic editorial detail

Create Oda Nobunaga's X post page (just before the Honnō-ji Incident).

[STYLE] monochromatic grayscale illustration, 3D rendered character, clean instructional reference sheet, white background, comic-style cell grid layout, technical diagram aesthetic [LAYOUT] 4x4 grid layout, 16 panels total, each panel separated by thin black border lines, numbered cells from 1 to 16, consistent panel size [CHARACTER] (enter character description) Example: young female dancer, athletic build, ponytail hairstyle, crop top and baggy pants, sneakers, same character in all panels [PANEL STRUCTURE - per cell] top-left: bold number badge + Korean title text center: full-body character pose illustration bottom-left: Korean description text (3-4 lines) overlay: directional arrows indicating movement direction [ARROWS / MOTION INDICATORS] curved arrows, straight arrows, circular rotation indicators, placed around the character to show movement flow and direction [RENDERING STYLE] high detail 3D sculpt style, soft studio lighting, subtle shadows, no color, grayscale shading, clean linework, game concept art quality [NEGATIVE] no background scenery, no color tones, no extra characters, no cluttered backgrounds

A beautiful woman looking at her phone on the subway; a candid photo.

More prompts load as you scroll

GPT Image 2 prompts guide and FAQ

GPT Image 2 prompts work best when you describe subject, framing, lighting, material, mood, and output intent in one compact instruction. Strong prompts reduce guesswork, make visual style easier to repeat, and help teams compare results across campaigns, product shots, portraits, and concept art.

This page is designed to help you study real prompt patterns before you write your own. You can scan successful examples, copy patterns that already work, and adapt each prompt for brand ads, editorial scenes, UI mockups, or reference-based edits.

How to write stronger prompts for GPT Image 2

  • Start with the core subject, camera angle, and scene goal, then layer in lighting, lens, composition, and texture details.
  • Effective prompts usually mention the constraints that change the final asset most: aspect ratio, background control, palette, text treatment, or whether the image should feel cinematic, clean, playful, or documentary.
  • Review several strong prompt examples side by side, keep the parts that consistently improve quality, and save reusable prompt patterns for repeated workflows.

Prompt FAQ

What makes GPT Image 2 prompts different from generic image prompts?

The best prompts are explicit about visual intent and production constraints. Instead of asking for a vague scene, they tell the model what should stay sharp, what should feel branded, and which details must not drift.

How long should a prompt be?

The best GPT Image 2 prompts are usually as short as possible but as specific as necessary. If a short prompt misses style, subject hierarchy, or composition, expand it only with details that change the image in a reliable way.

Can prompts handle edits, references, and variations?

GPT Image 2 prompts are useful for fresh generations, reference-led edits, and controlled variations. When you want consistent output, keep the stable visual anchors the same and change only one or two variables per iteration.