GPT Image 2 Prompts Library · Halaman 58

A highly realistic AAA video game screenshot depicting a boss fight inside a dusty, golden-lit {argument name="setting" default="ancient Egyptian temple"}. In the center background, a massive {argument name="boss name" default="ANUBIS"} boss monster with a jackal head, glowing eyes, and ornate gold armor roars while wielding a staff, kicking up swirling sand. In the left foreground, a muscular, bald warrior with red body tattoos faces the boss, holding a {argument name="weapon" default="curved blade"}. In the right midground, a young boy aims a bow at the boss. The image features 4 distinct UI components: a top-center red boss health bar labeled '{argument name="boss name" default="ANUBIS"}', a right-aligned quest tracker labeled '{argument name="quest name" default="THE JACKAL'S PATH"}' with the subtext 'Defeat Anubis', a bottom-left player status HUD showing health and rage bars alongside 4 diamond-shaped skill icons with button prompts, and a bottom-right companion HUD showing 2 circular skill icons.

Please create an X (Twitter) post page for Oda Nobunaga ({argument name="situation" default="just before the Honno-ji Incident"}).

A realistic, slightly grainy, low-lighting mugshot of a young woman with long, messy black hair, prominent cheekbones, and full lips. She has a small tattoo near her right eye and extensive, intricate mandala-style tattoos covering her neck, chest, and both hands/arms. She is wearing a simple black sleeveless top. She is holding up a rectangular white placard with blurred-out text, held with both hands showing long, manicured nails. The background is a plain white cinderblock wall with a height chart on the right side, showing numbers like 150, 160, and 170 in black stencil font. The overall aesthetic is gritty and raw, resembling a genuine police booking photograph.

A photorealistic street style paparazzi shot of a young woman walking through a {argument name="setting" default="sunny parking lot"}. She has {argument name="hair color and style" default="blonde hair in a high ponytail"} and is wearing {argument name="accessory" default="black rectangular sunglasses"}. Her outfit consists of a matching {argument name="activewear color" default="grey"} long-sleeve crop top and high-waisted leggings, paired with chunky white sneakers and white crew socks. Over her left arm, she carries a light blue denim jacket, while holding a {argument name="held item" default="beige insulated water bottle"} and a smartphone in the same hand. The background is blurred with a shallow depth of field, showing the rear of a dark car on the left and out-of-focus buildings and trees in the distance, capturing a casual, candid athleisure aesthetic in bright daylight.

{
  "type": "two-page yoga instruction book spread",
  "subject": "Asian woman with hair in a high bun, wearing {argument name=\"subject outfit\" default=\"blue sports bra and matching leggings\"}",
  "layout": {
    "left_page": {
      "header": {
        "title_english": "{argument name=\"pose name english\" default=\"Eka Pada Koundinyasana I\"}",
        "subtitle_english": "One-Legged Sage Koundinya's Pose I",
        "title_korean": "{argument name=\"pose name korean\" default=\"에카 파다 콘디니아사나 I\"}",
        "subtitle_korean": "한 다리로 균형을 잡는 현자의 자세 I"
      },
      "main_image": "large photo of the subject performing the full arm balance pose on a grey yoga mat",
      "description": "paragraph of Korean text explaining the pose",
      "bottom_columns": [
        { "title": "효과", "count": 3, "description": "list of physical benefits" },
        { "title": "준비 자세", "count": 3, "description": "list of preparatory poses" }
      ],
      "footer": "page {argument name=\"page number left\" default=\"102\"} and section title"
    },
    "right_page": {
      "header": "HOW TO DO / 따라 해보기",
      "steps_grid": {
        "count": 4,
        "description": "numbered steps 1 to 4 with Korean instructions and progression photos",
        "images": [
          "Step 1: sitting, preparing arm and leg",
          "Step 2: leaning forward, hands on mat",
          "Step 3: lifting off the ground",
          "Step 4: full balanced pose"
        ]
      },
      "tip_box": { "title": "TIP", "count": 3, "description": "bulleted Korean text lines in a shaded box" },
      "footer": "section title and page {argument name=\"page number right\" default=\"103\"}"
    }
  }
}

{
  "type": "retro pixel art video game mockup",
  "theme": "survival death game",
  "hud": {
    "top_left": ["HIGH SCORE", "000000"],
    "top_center_title": "{argument name=\"game title\" default=\"오징어 게임\"}",
    "top_right": ["PLAYER 001", "3 red hearts"]
  },
  "menu": {
    "style": "centered under title, black boxes, white pixel font, cursor pointing to first item",
    "items": [
      "{argument name=\"menu item 1\" default=\"게임 시작\"}",
      "{argument name=\"menu item 2\" default=\"설정\"}",
      "{argument name=\"menu item 3\" default=\"종료\"}"
    ]
  },
  "gameplay_ui": {
    "mid_left_panel": ["ROUND 1", "{argument name=\"round description\" default=\"무궁화 꽃이 피었습니다\"}"],
    "mid_right_panel": ["01:00 digital red timer"]
  },
  "scene": {
    "background": "blue sky, pixel clouds, 2 small houses, tall yellow grass field",
    "midground": "large bare tree, 1 giant doll in yellow shirt and orange dress facing the tree, 2 guards in pink jumpsuits with square masks holding rifles",
    "foreground": "pink starting line on dirt ground, 9 players in green tracksuits facing away from the viewer",
    "player_numbers": ["019", "240", "199", "218", "001", "067", "456", "101", "101"]
  }
}

A photorealistic underwater portrait of a {argument name="subject description" default="young Asian woman"} holding her breath just below the surface. She has {argument name="hair style" default="long dark wavy hair"} floating weightlessly around her face and is wearing a {argument name="clothing" default="sheer white ruffled dress"}. She looks directly at the camera with slightly puffed cheeks and pursed lips. The scene is illuminated by bright sunlight filtering through the {argument name="environment" default="clear blue water"}, casting intricate, dappled caustic light patterns across her face, chest, and clothing. Small, delicate air bubbles float upward around her face, adding to the serene, ethereal atmosphere.

Amateur photo of {argument name="item" default="an open notebook"} lying flat, filled with handwritten notes in {argument name="pen type" default="black ballpoint pen"}. The handwriting is casual and slightly messy, like personnal notes, natural imperfections, crossed out words, underlined headings. Shot from slightly above, natural daylight from a window, no flash. Casual desk setting, shot on {argument name="device" default="iPhone"}

A full-body portrait of a person standing in a {argument name="setting" default="narrow wet cobblestone alleyway at night"}. The person is wearing a {argument name="clothing" default="dark jacket and dark trousers"}, with one hand resting against the rough, textured wall on the left and the other hand casually tucked into their pocket. Their face is completely obscured by a {argument name="face replacement" default="large yellow smiling emoji with blushing cheeks"}. The scene is illuminated by a single vintage street lamp mounted on the right wall, casting warm yellow light that reflects beautifully off the wet paving stones, creating deep, moody shadows. In the background, a small illuminated sign reads "{argument name="background sign text" default="三条小路"}". The artwork is a {argument name="art style" default="modern digital illustration with high contrast lighting and subtle shin-hanga influences"}, featuring a red square seal in the bottom right corner with the text "新版画" and a blank white vertical signature box with a tiny red seal in the bottom left.

An anime-style illustration of a figure leaning against a textured stone wall in a narrow, sunlit alleyway. The figure has a {argument name="head type" default="large yellow smiling emoji with rosy cheeks"} for a head. They wear a {argument name="outfit" default="loose black overcoat, black shirt, wide-leg black trousers, and black dress shoes"}. The character's left foot rests on a stone block, right hand in pocket, and left hand near the neck. The scene includes reflective stone pavement, a distant streetlamp, and golden-lit foliage. The lighting features dramatic chiaroscuro and sharp shadows.

A realistic image shot with {argument name="lighting style" default="harsh direct flash photography"} inside an {argument name="environment" default="ice cave made of large ice blocks and snow"}. A young man with messy dark curly hair sits on a brown fur pelt, leaning his head on his hand with a relaxed expression. He is wearing a {argument name="clothing" default="thick black and white patterned knit sweater"}, black snow pants, and black winter boots. Directly behind him, the rear half of a white {argument name="car model" default="Porsche 911 GT3 RS"} sports car is embedded into the snowy cave wall, with its red taillights illuminated and a license plate reading {argument name="license plate" default="VS 369 911"}. In the foreground, a rough wooden table dusted with snow holds exactly 2 stainless steel thermoses, 1 clear textured glass, 1 small metal cooking pot, and 1 yellow and blue matchbox. Resting on the icy floor to the right are 1 green glass bottle and 1 vintage metal lantern.

First-person perspective video game screenshot showing hands wearing {argument name="glove style" default="Spider-Man suit"} gloves holding an AK-47 assault rifle. The weapon features a custom {argument name="weapon theme" default="Spider-Man"} skin with a red and blue color scheme, black webbing patterns, a white spider emblem on the receiver, "CS2" on the handguard, "MARVEL" on the magazine, and "SPIDER-MAN" on the stock. The background is a dusty Middle Eastern town street resembling a tactical shooter map, featuring a sign reading "HOTEL AURORE" with an arrow pointing to "A". The top center UI displays a red shield with a black spider, the text "{argument name="weapon title" default="AK-47 | Web-Slinger"}", and "The CS2 Collection" above a red line. The bottom UI features exactly 5 minimalist white icons on the left and the word "CLOSE" on the right.

A candid, photorealistic portrait of a young woman with {argument name="hair color" default="long blonde hair"} wearing a {argument name="outfit" default="dark brown leather blazer"}. She is sitting at a table in a {argument name="setting" default="restaurant or wine bar"}, resting her chin on her hand and looking off to the side with {argument name="expression" default="pursed lips"}. In front of her on the white tablecloth are two glasses: a large glass of {argument name="drink" default="red wine"} and a smaller glass of water. The background features brightly lit, modern shelving filled with rows of wine bottles. Warm, ambient lighting, shot on a smartphone.

{
  "type": "two-sided event flyer mockup",
  "style": "watercolor illustration, warm earthy tones, cozy, inviting, rustic",
  "left_side_front": {
    "header": {
      "title": "{argument name=\"event name\" default=\"SENDAI COFFEE PICNIC 2026\"}",
      "subtitle": "{argument name=\"subtitle\" default=\"街のカフェが集まる、春の1日\"}",
      "tagline": "A relaxing time with your favorite cup"
    },
    "main_illustration": "People enjoying a picnic on a grassy field. A family sitting on a checkered blanket drinking coffee. Background features wooden coffee stalls, tents, trees, and a distant city skyline.",
    "event_info": {
      "date": "{argument name=\"event date\" default=\"2026年5月16日(土) - 17日(日)\"}",
      "time": "10:00 - 17:00",
      "location": "{argument name=\"location\" default=\"七北田公園 芝生広場\"}",
      "badge": "Orange oval badge with text for free entry"
    },
    "footer": {
      "sponsor_logos": 4,
      "elements": ["QR code", "grass and white flowers border"]
    }
  },
  "right_side_back": {
    "header": {
      "title": "{argument name=\"event name\" default=\"SENDAI COFFEE PICNIC 2026\"}",
      "illustration": "small green tree"
    },
    "sections": [
      {
        "title": "Vendor List",
        "layout": "2 columns",
        "count": 9,
        "labels": ["Blue Ridge Roastery", "NORTH BEANS LAB", "SUNNY DAY COFFEE", "KITSUNE COFFEE WORKS", "Forest Table Cafe", "creme & crumb", "GOOD DAY SAND", "HOSHINO KITCHEN CAR", "PIZZA VAN 88"],
        "illustrations": ["sandwiches", "pizza", "coffee cups"]
      },
      {
        "title": "Event Timetable",
        "layout": "2 columns",
        "count": 7,
        "description": "Schedule with times like 10:30, 11:30, 13:00, 14:30, 16:00 and event names"
      },
      {
        "title": "Feature Highlights",
        "layout": "3 horizontal boxes",
        "count": 3,
        "labels": ["Latte Art Experience", "Coffee Tasting Stamp Rally", "Food & Sweets Area"],
        "illustrations": ["single coffee cup", "multiple coffee cups", "food basket"]
      }
    ],
    "footer": {
      "social_media": "{argument name=\"social handle\" default=\"@sendai_coffee_picnic\"}",
      "icons": ["Instagram logo"],
      "elements": ["QR code", "brown background strip"]
    }
  }
}

Modern pencil illustration of {argument name="location" default="Vintage travel poster illustration of the Amalfi Coast, Italy"}, panoramic coastal cliff road scene, classic 1960s white car driving along a curved seaside road, deep blue Mediterranean sea with small sailboats, colorful pastel hillside village, bright blue sky with soft clouds, lemon tree branches with vibrant yellow lemons framing the foreground, warm summer sunlight, bold vibrant colors, {argument name="style" default="retro 1950s travel poster"} style, cinematic composition, high detail, screen print texture, graphic illustration. Hand-drawn style, illustration with loose strokes and defined contours. High-contrast color palette, maintaining chromatic harmony between background and elements. Contemporary and decorative aesthetic.

{
  "type": "photograph of an open notebook",
  "setting": "wooden desk surface, top-down angle",
  "subject": {
    "hands": "two hands visible at the bottom, wearing a dark blue ribbed sweater, right hand holding a red pen labeled {argument name=\"pen brand\" default=\"PILOT G-TEC-C4\"} writing on the page",
    "notebook": "open A5 dot-grid notebook with handwritten notes and diagrams in black, blue, green, and red ink"
  },
  "layout": {
    "left_page": {
      "header": "{argument name=\"main title\" default=\"「GPT-Image-2」による3Dモデル三面図生成の検証\"}",
      "sections": [
        {
          "title": "1. 検証フロー (Workflow):",
          "content": "text flow and a flowchart diagram",
          "flowchart_elements": 5,
          "flowchart_labels": ["Input (Prompt)", "GPT-Image-2", "正面図", "側面図", "背面図"]
        }
      ]
    },
    "right_page": {
      "sections": [
        {
          "title": "2. 結果分析",
          "target_box": "Target [{argument name=\"target object\" default=\"椅子\"}] => [3D Model & 3面図 images]",
          "bullet_points_count": 3,
          "bullet_point_topics": ["精度: ★★★☆☆", "整合性", "課題"],
          "diagram_count": 3,
          "diagram_labels": ["正面", "側面", "背面"],
          "diagram_description": "sketches of chairs with red correction notes on the front view"
        }
      ],
      "bottom_right_corner": {
        "doodle": "cute anime-style line drawing of a smiling girl with short hair",
        "label": "{argument name=\"character name\" default=\"あいちゃん\"} {argument name=\"date\" default=\"3/15\"}"
      }
    }
  }
}

A photograph of a tech keynote presentation. A bald man with glasses, wearing a dark blue sweater and jeans, stands on a stage to the left, smiling and pointing with both index fingers at a large presentation screen behind him. The screen displays a slide with a light blue background. At the top left is a small teal pill-shaped badge reading "NEW DATA". Below it, the main headline reads "{argument name="headline" default="Microsoft Edge is the #1 browser to download Google Chrome"}", with "#1" highlighted in blue. A subheadline reads "% of Google Chrome downloads that started from another browser". Below this is a bar chart with three bars. The first bar is tall and blue, labeled "38.6%" above it, with the Microsoft Edge logo and text "Microsoft Edge" below it. The second bar is shorter and gray, labeled "17.2%" above it, with the Mozilla Firefox logo and text "Mozilla Firefox" below it. The third bar is the shortest and gray, labeled "13.0%" above it, with the Apple Safari logo and text "Apple Safari" below it. At the bottom left of the slide, small text reads "Source: Microsoft internal analysis of Google Chrome download referrals Cohort: Windows 10 and 11, Worldwide, Jan - March 2024". To the left of the screen, a wooden panel features the Microsoft logo (four colored squares). The backs of audience members' heads are visible in the foreground at the bottom.

A wide image shot on {argument name="camera" default="iPhone"} of a glass whiteboard, in a room overlooking {argument name="landmark" default="the Bay Bridge"}. The field of view shows {argument name="subject" default="a woman writing"}, sporting a tshirt with a large OpenAl logo. The handwriting looks natural and a bit messy, and we see the photographer's reflection.

A photorealistic candid photograph of {argument name="person" default="Sam Altman"} working behind the counter at a {argument name="restaurant brand" default="McDonald's"} restaurant. He is wearing a dark grey short-sleeve t-shirt under a black apron with yellow neck straps and the restaurant's logo, along with a white nametag reading "{argument name="nametag text" default="Sam"}". He has a slightly tired, focused expression, holding a {argument name="held item" default="red carton of french fries"} in his left hand while extending his right hand forward to serve a {argument name="served item" default="white drink cup with a straw"}. The background features a realistic, slightly blurred fast-food kitchen environment with stainless steel counters, a fry station glowing under warm heat lamps, digital order screens, a point-of-sale register, and another employee working in the distance. The lighting is typical indoor fluorescent, giving the image a documentary, everyday feel.

A first-person POV screenshot inside a {argument name="game environment" default="VRChat lobby"} featuring a dark checkered floor, a "VR CHAT" welcome sign on the left, a "Mirror" doorway, and an "Avatar Showcase" on the right. The viewer's hands are visible in the foreground, wearing white sleeves with red dashed stitching, reaching forward. A red muted microphone UI icon sits in the bottom left corner. In the center stands {argument name="tall character" default="Ran Mouri from Detective Conan in her blue school uniform"}, smiling with closed eyes and {argument name="action" default="patting the heads of the two smaller characters"}. On her left is {argument name="small character 1" default="Conan Edogawa in his blue suit and red bowtie"}. On her right is {argument name="small character 2" default="the Ugandan Knuckles meme character"}. Each character has a floating nameplate above them: the tall girl's reads "Ran Mouri" in yellow over "LetsAskAQuestion" in white, the boy's reads "Conan Edogawa" over "conan.exe", and the red creature's reads "Ugandan Knuckles" over "do u know de way".

Amateur iPhone photo of {argument name="subject" default="Kylian Mbappé"} in tears, wearing a {argument name="team jersey" default="Real Madrid jersey"} on the pitch, looking devastated. In the background, {argument name="opponent players" default="Bayern Munich players in red kits"} are visible. Shot from the stands close to the field, at night inside the {argument name="stadium" default="Allianz Arena"}

Amateur candid photo taken in an {argument name="time period" default="early 2020s"} {argument name="location" default="high school"}, {argument name="aesthetic" default="iPhone aesthetic"}

{
  "type": "web browser screenshot",
  "interface": "Google search results page",
  "header": {
    "logo": "Google",
    "search_bar": {
      "query": "{argument name=\"search query\" default=\"openai gpt-image 2 image generator\"}",
      "icons": ["clear", "microphone", "camera", "search"]
    },
    "right_actions": ["settings gear", "apps grid", "blue Sign in button"]
  },
  "navigation_tabs": ["All", "Images", "News", "Videos", "Shopping", "Short videos", "Forums", "Web", "More", "Tools"],
  "results_stats": "About 18,200,000 results (0.32 seconds)",
  "search_results": {
    "count": 3,
    "items": [
      {
        "source": "OpenAI",
        "url": "https://openai.com > index > gpt-image-2",
        "title": "{argument name=\"result 1 title\" default=\"GPT-Image-2 | OpenAI\"}",
        "snippet": "{argument name=\"date\" default=\"Apr 1, 2025\"} — GPT-Image-2 is our new image generation model that delivers state-of-the-art image quality, finer details, better text rendering, and more precise instruction following.",
        "sitelinks": ["Research", "Safety", "Try in ChatGPT"]
      },
      {
        "source": "OpenAI Platform",
        "url": "https://platform.openai.com > docs > gpt-image-2",
        "title": "GPT-Image-2 - OpenAI API",
        "snippet": "Learn about GPT-Image-2, OpenAI's most advanced image generation model, available via the Images API. Create high-quality images with improved realism, text rendering, and instruction ...",
        "sitelinks": ["Quickstart", "Edit images", "Example code"]
      },
      {
        "source": "The Verge",
        "url": "https://www.theverge.com > openai-gpt-image-2-launch",
        "title": "OpenAI's GPT-Image-2 brings major improvements to AI ...",
        "snippet": "Apr 1, 2025 — OpenAI has officially launched GPT-Image-2, its next-generation image generator that offers better photorealism, improved text rendering, and more accurate adherence to user ..."
      }
    ]
  },
  "people_also_ask": {
    "title": "People also ask",
    "count": 4,
    "questions": [
      "What is GPT-Image-2?",
      "How is GPT-Image-2 different from previous models?",
      "Can I try GPT-Image-2 in ChatGPT?",
      "Is GPT-Image-2 available through the API?"
    ]
  }
}

{
  "type": "wikiHow article mockup",
  "header": {
    "background": "green",
    "elements": [
      "hamburger menu icon",
      "wikiHow logo",
      "search bar with 'Search how to...' text",
      "5 navigation icons: PRO/PPRO, QUIZZES, EDIT, EXPLORE, LOG IN"
    ]
  },
  "breadcrumbs": "HOME » CATEGORIES » COMPUTERS AND ELECTRONICS » INTERNET » CHATGPT",
  "article_info": {
    "title": "{argument name=\"article title\" default=\"How To Use GPT-Image 2\"}",
    "metadata": "Written by wikiHow Staff | Updated: June 10, 2024",
    "sidebar": [
      "PDF Download Article button",
      "METHODS box: 1 Using GPT-Image 2 on Arena.ai"
    ]
  },
  "intro_box": "GPT-Image 2 is OpenAI's most advanced AI image generator. It creates high-quailly, detailed, and accurate images from your text prompts. This wikiHow will show you how to try GPT-Image 2 for free on arena.ai.",
  "steps": [
    {
      "number": 1,
      "title": "{argument name=\"step 1 title\" default=\"Go to arena.ai.\"}",
      "description": "Open your web browser and navigate to https://arena.ai. Arena is a platform where you can compare and try thousands of AI models, including OpenAI's latest image generator.",
      "embedded_image": {
        "type": "dark mode website UI mockup",
        "logo": "arena",
        "navigation": ["Battle", "Side-by-Side", "Direct Chat", "Leaderboard", "Community", "Sign In", "Sign Up button"],
        "hero_text": "The world's largest arena for AI models",
        "input_ui": {
          "model_dropdown": "Model: {argument name=\"model name\" default=\"duct-tape-3\"}",
          "settings": "Default",
          "text_input": "Enter your prompt here...",
          "button": "Run"
        }
      }
    },
    {
      "number": 2,
      "title": "{argument name=\"step 2 title\" default=\"Select the model 'duct-tape-3.'\"}",
      "description": "Click the Model dropdown menu and choose 'duct-tape-3.' Duct-tape-3 is the name Arena uses for GPT-Image 2, OpenAI's most capable image generation model."
    },
    {
      "number": 3,
      "title": "{argument name=\"step 3 title\" default=\"Prompt it.\"}",
      "description": "Type your prompt in the box and click Run. GPT-Image 2 will generate an image based on your description.",
      "tip_box": "TIP GPT-Image 2 is the best AI image generator because it produces incredibly detailed, accurate, and high-quality images, handles text well, follows complex instructions, and has a deep understanding of the world."
    }
  ]
}