AI 提示词库

Ultra-real cinematic night view of a modern city where the invisible electrical power grid becomes faintly visible in the air, subtly connecting buildings and streets under realistic lighting and grounded film realism.

Please make it not look like AI! Since I get asked this, I created Nano Banana Pro prompts that aren't obviously AI-generated. Vague instructions are OK, and they are reproducible! For example, this was generated just by inputting the theme: "{argument name="theme" default="Design x AI Strategy"}"

Deep gradient background with subtle atmospheric fade. Foreground soft blur, midground clarity, background fade into light, muted tonal transitions, cinematic depth feel.

If it's 'Good morning' it appears, but 'After this' doesn't come out properly.

add a playhouse behind mickey

Change the chair in front of the bathtub in the first image to the one in the second image. Please adjust the size and light source so that it blends in without any sense of incongruity.

Geometric Clay Figures

{argument name="helmet type" default="3D printed Corinthian netrunner helmet"} where the plume is a heatsink

Mozart From Requiem in D minor, K.626 {argument name="piece name" default="Dies Irae"}

一张宽幅引言卡片，上面印有一位名人，背景为棕色，引言文字为浅金色衬线字体：“{argument name="famous_quote" default="Stay Hungry, Stay Foolish"}”，下方是较小的文字：“—{argument name="author" default="Steve Jobs"}”。文字前面有一个大而柔和的引号。人物肖像在左侧，文字在右侧。文字占据图片的三分之二，肖像占据三分之一，肖像部分带有轻微的渐变过渡效果。

can let it explode, static

Input this instruction, allowing the AI to complete the entire process step-by-step: extracting selling points, determining style, determining layout, and finally generating a complete, system-wide prompt. Finally, give the AI-generated prompt to the Nano Banana Pro model (it must be Pro). You can first provide some overall style prompts separately as general constraints. Then, whether generating images individually or in batches using third-party tools, it is possible.

location: description: > Simple background created by interpreting the reference image’s mood. Use soft gradients, atmospheric glow, light particles, or abstract light rays. Avoid complex textures or any copyrighted shapes. background_feel: > Subtle, atmospheric, clean.

Make the previous image into a single subject like the next image

Ultra-high-quality hyper-realistic photo featuring a real human and a giant 3D Pixar-style version of THE SAME person — both strictly based on the uploaded reference image. The real person and the Pixar character must have identical facial structure, hairstyle, facial expression, body proportions, height, and clothing. Both are wearing a knitted sweater with a gradient from light yellow to pink, high-waisted blue jeans, and white high-top sneakers. The pose must exactly match the reference: the real person stands relaxed with a friendly smile, with their hand naturally resting on the shoulder of the Pixar character. The Pixar version stands slightly wider, arms confidently positioned near the pockets, with a playful sly smirk, one raised eyebrow, and expressive Pixar-style eyes. The Pixar character should be proportionally larger, but anatomically identical to the real person. The facial expression, head tilt, posture, and body language must precisely replicate the reference image. Clean gray-blue studio background, soft even studio lighting, realistic shadows, crisp fabric texture, smooth Pixar-style skin shading, cinematic sharpness. DO NOT change the pose, reaction, clothing, or proportions. Strict adherence to the reference.

A {argument name="character type" default="[CHARACTER / FIGURE]"} entirely built from flowing {argument name="energy type" default="[ENERGY / LIGHT / COLORS]"}, as if captured in motion. Trails form limbs and features, glowing softly in {argument name="atmosphere" default="[ATMOSPHERE]"}, dramatic cinematic composition.

Use the same face from the reference image without changing facial features { "subject": { "gender": "female", "features": "doe-eyed, full lips with pink gloss, high-arched eyebrows, soft facial contour", "hair": "voluminous {argument name="hair color" default="strawberry blonde"}, wavy, tousled texture, wind-swept", "makeup": "soft-glam, shimmering lilac eyeshadow, inner corner highlight, heavy rosy blush on cheeks and nose, fluttery eyelashes", "pose": "reclined, hand gently touching chin, direct gaze at camera, relaxed posture", "tattoos": "small crescent moon on left shoulder" }, "fashion_and_accessories": { "clothing": "pink sequined tube dress, thin beaded straps, shimmering texture", "jewelry": [ "multiple stacked gold bangles on both wrists", "thin silver band rings", "delicate gold necklace" ] }, "environment": { "setting": "tropical beach at night", "background_elements": ["palm tree trunk", "palm fronds", "white sand", "dark night sky"], "lighting": "soft frontal flash, cool-toned ambient night light, highlighting skin texture and sequins" }, "technical_style": { "medium": "high-resolution digital photography", "aesthetic": "social media influencer style, soft focus, dreamy, ethereal", "color_palette": ["rose pink", "gold", "sandy beige", "midnight blue", "soft lilac"], "composition": "medium close-up, slightly high-angle shot" } }

{ "prompt_description": "A detailed generation prompt for a cinematic winter portrait based on the provided image reference.", "subject": { "age_group": "Young adult woman, mid-20s", "expression": "Calm, gentle, confident smile looking directly at the camera", "features": "Natural beauty, long dark wavy hair illuminated by backlight", "pose": "Standing medium shot, hands casually tucked into coat pockets" }, "apparel": { "coat": "Thick, textured dark charcoal grey wool double-breasted pea coat", "hat": "Black ribbed knit beanie pulled snugly onto head" }, "environment": { "setting": "Deep winter pine forest at twilight", "elements": "Heavy snow cover on ground and trees, actively falling large snowflakes", "background": "Blurred snow-laden pine trees creating a cool blue backdrop" }, "lighting": { "style": "Dramatic cinematic backlighting (rim light)", "details": "Strong warm golden light source behind the subject creating a brilliant halo outline on her hair, shoulders, and the falling snow, contrasting with cool blue ambient evening light" }, "photography_style": { "type": "Cinematic portrait photography", "quality": "Ultra-realistic, 8k resolution, highly detailed textures", "lens_effects": "Shallow depth of field (strong bokeh in background), sharp focus on subject", "color_grade": "Teal and orange contrast, emphasizing warm highlights against cool shadows" } }

a banana, hyper detailed, photorealistic, 8k, professional photo, high quality, high resolution, --ar 16:9 --style raw --v 6.0

A {argument name="setting" default="quiet library"} and a {argument name="style" default="bright red traditional look"}. A love for books & style.

Create and output one picture {argument name="style" default="like Vermeer"}!

A photorealistic image of a leaf that looks like it’s in the shape of a lion.

A hyper-detailed cinematic portrait of a human face use selected image for face reference in side profile, looking upward with a calm and powerful expression, the entire subject illuminated in deep {argument name="light color" default="electric blue"} light, water splashes and liquid energy wrapping around the head and shoulders, high contrast, dramatic lighting, ultra-realistic skin texture, fine pores visible, sharp focus, 8K resolution, volumetric lighting, dark blue background, futuristic, ethereal, emotional, digital art, photorealism, high detail.

A surreal, photorealistic scene of a tiny adult man crouching barefoot on a green plastic bottle cap, cupping his hands to drink water pouring from the mouth of an enormous green plastic bottle. The bottle lies on its side atop a mossy stone ledge, releasing a steady stream of clear water. The setting is outdoors on a wet stone pathway with shallow puddles, soft reflections, and lush green trees blurred in the background. Cinematic depth of field, natural daylight, ultra-detailed textures, realistic water droplets, whimsical scale contrast, high realism, 8k quality.