Which AI model should I use?

Start with Nano Banana 2, which is the model associated with this template. You can compare another compatible model for a different finish.

High-End Commercial Product Photography for an Energy Drink — AI Image Template for nano-banana-2

Satoru Gojo Anime Propaganda Poster

Create a vertical Japanese anime propaganda poster for {argument name="character name" default="Satoru Gojo"} from Jujutsu Kaisen, with a consistent hierarchical main visual composition and dramatic ink-wash collage style. Canvas: portrait 3:4 poster on warm off-white aged paper with subtle grain, dust specks, faint scratches, and scattered ink splatters. Visual hierarchy: the top half is dominated by an enormous right-facing side profile of Gojo wearing a black blindfold, smiling slightly, with spiky white hair exploding upward into translucent feather-like shards and blue-purple energy strokes. In the center foreground place a full-body Gojo walking toward the viewer, tall and slim in a high-collar black uniform, blindfolded, one hand in his pocket and the other raised with two fingers extended; red cursed-energy lightning coils around his torso and legs. Use deep indigo, violet, blue, black, white, and crimson accents, with painterly splashes and sharp manga linework. Composition details: include exactly 6 smaller character/story vignettes blended into the ink cloud around the main figure: 1) a blue-haired male silhouette near the upper left within the hair/energy mass, 2) a white-haired blindfolded smaller Gojo in profile at upper left, 3) a dark standing figure facing the central energy on the mid-right, 4) a small lone silhouette in the lower-left mist, 5) a group of three small figures walking in the lower middle background, and 6) a red-haired/pink-haired male with a black blindfold or markings in the lower-right brush splash. Add exactly 4 environmental motifs: 1) a traditional Japanese temple building on the mid-right, 2) a modern city skyline at bottom center, 3) a red torii gate at the lower foreground behind the central figure’s legs, and 4) tiny bird silhouettes in the white margins. Keep all secondary elements partially dissolved into blue and purple watercolor smoke. Text layout: on the left side, large vertical Japanese brush-calligraphy reading {argument name="large Japanese title" default="五条悟"}, painted in dark navy ink. Beside it add small vertical roman text reading “SATORU GOJO” and a small red seal with Japanese characters. On the right side, add clean vertical Japanese text reading {argument name="right side slogan" default="最強の呪術師 — 五条悟"}, with a small circular emblem and the Japanese series title beneath it. On the lower left, add smaller vertical Japanese text reading {argument name="lower left quote" default="大丈夫僕最強だから"}, mixing black and muted red characters. Overall mood: heroic, mythic, national-poster-like propaganda energy without real-world politics, elegant manga poster design, high detail, strong central silhouette, layered ink splashes, luminous cursed energy, crisp face profile, no English headline except the small “SATORU GOJO,” no watermark, no extra characters beyond the listed vignettes.

$Dramatic Chibi Anime Poster — image_generation_prompt: meta: description: "{argument name="シーン説明" default="ちびキャラがロードバイクに乗ろうとしている後ろ姿のアニメ映画ポスター風イラスト"}"$

Dramatic Chibi Anime Poster

image_generation_prompt: meta: description: "{argument name="シーン説明" default="ちびキャラがロードバイクに乗ろうとしている後ろ姿のアニメ映画ポスター風イラスト"}" reference_usage: "キャラクターの見た目は添付画像を参照" style: art_style: "{argument name="アートスタイル" default="アニメ映画のポスター (Anime movie poster)"}" mood: "ドラマチック" subject: type: "ちびキャラ (Chibi character)" pose: view: "後ろ姿 (Back view)" action: "画面奥（真上）のロードバイクに向かって左足を踏み出している" target_direction: "ロードバイクの方向（画像の真上方向）" composition: top: element: "赤色のロードバイク (Red Road bike)" layout: "画面の横幅いっぱいの大きさ、画像上端と平行" center: element: "ちびキャラの後ろ姿" angle: "俯瞰 (Overhead)" bottom: text_layer_base: content: "{argument name="テキスト" default="KASANE"}" style: "bold sans-serif font" color: "白色 (White)" text_layer_overlay: content: "カサネ" style: "太い筆でなぐり書きしたような荒々しい文字" color: "赤色 (Red)" position: "KASANEの文字の上に被る形" objects: road_bike: color: "赤色 (Red)" orientation: "真横 (Side view / Horizontal)" facing_direction: "左側 (Facing left)" road: detail: "道路の中央にまっすぐな白線が引かれている。その白線は、はっきりとは見えず、薄っすらと、ぼんやりとしか見えない。道路の左側の一部分に、かすかに、薄っすらと、ほんのすこしだけひび割れがある。" technical_settings: shot_size: "ロングショット (long shot)" shot_angle: "俯瞰 (overhead shot)" zoom_emphasis: "ちびキャラと自転車を大きく強調" lens_effects: "ビネッティング (Vignetting)" lighting: direction: "真上から照射 (From above)"

3x3 Grid Eye Expression Test Prompt (Right Image)

A single square image divided into a clean 3x3 grid of nine equal squares. Each square contains the girl character from image1, close-up portrait, head and neck only, face filling most of the frame. Each square has a different eyes. Place the Japanese text below the each square as a label, not touching the edges. 1：螺旋状の目: {argument name="目1" default="spiral-only eyes"} 2：空白な目: {argument name="目2" default="blank eyes"} 3：虚ろな目: {argument name="目3" default="hollow eyes"} 4：腫れた目: puffy eyes 5：不気味な目: creepy eyes 6：狂気の目: crazy eyes 7：誘惑的な目: mesmerizing eyes 8：色っぽい目: sensual eyes 9：機械の目: mechanical eyes Simple blue background colors. High resolution. Sticker sheet layout, perfect 3x3 alignment.

gpt2-profile-avatar-restored-vintage-mother-and-child-portrait

A restored vintage family snapshot, photographed indoors in soft natural light, showing a {argument name="adult subject" default="young mother"} seated and holding a {argument name="child subject" default="toddler"} on her lap in a close, centered waist-up portrait. The adult has short softly curled auburn hair in a voluminous 1960s-inspired bob, wears a sleeveless black dress and a thin gold necklace, and wraps both arms protectively around the child. The child has fine light blond hair and wears a plain white long-sleeve outfit. Compose the image with a warm nostalgic color cast, gentle film softness, subtle grain, and the look of a carefully repaired old printed photograph. Place them in front of a cream-colored curtain patterned with small brown teddy bear motifs, with a softly blurred interior window frame visible along the top background. Preserve realistic skin tones, natural posture, and the intimate family-photo feeling, as if an old damaged photograph has been professionally reimagined and restored. Square crop, centered composition, shallow depth of field, authentic analog photo texture, no modern styling, no text.

Identity-Based Rose Garden Portrait

Use my uploaded portrait only as the identity reference for the person’s facial features, face shape, hairstyle, hair color, skin tone, and overall real-person appearance. Do not copy the face or identity from the reference image. Recreate the uploaded person in a {argument name="setting" default="romantic rose-garden portrait"} inspired by the reference photo. The person is seated gracefully on a pale stone garden path, surrounded closely by dense blooming pink roses, soft green leaves, and layered floral bushes. Use a natural seated pose with the legs folded comfortably to one side, relaxed shoulders, elegant hands, and a calm, gentle expression while looking toward the camera. Choose an outfit automatically based on the uploaded person’s appearance, hairstyle, body proportions, and personal style. The clothing should feel flattering and harmonious with the romantic garden setting, using soft {argument name="colors" default="blush pink, ivory, cream, or pastel tones"}, with tasteful floral, chiffon, lace, ribbon, or layered fabric details. Do not force one specific dress design. Keep the outfit realistic, refined, and physically believable. Soft diffused daylight, subtle backlight through the flowers, delicate shadows, dreamy pastel color grading, shallow depth of field, softly blurred background roses, romantic editorial photography, realistic fabric texture, natural hair strands, realistic skin pores, accurate anatomy, natural hands and fingers, lifelike body proportions, high facial identity accuracy, professional full-body portrait, square composition. Keep the uploaded person clearly recognizable and photorealistic. No artificial beauty filter, no excessive skin smoothing, no plastic skin, no doll-like face, no anime or illustration style, no CGI appearance, no watermark, no logo, no text. Negative Prompt： anime, manga, illustration, 3D render, CGI, doll face, plastic skin, waxy skin, excessive retouching, over-smoothed face, fake facial features, identity change, copied reference face, distorted anatomy, incorrect body proportions, extra fingers, fused fingers, malformed hands, stiff pose, floating limbs, overly revealing outfit, exaggerated accessories, oversaturated pink, harsh lighting, blurry face, low resolution, watermark, logo, text, signature

PRO

NewJeans Danielle Portrait Prompt for Nano Banana Pro

{ "prompt": "{argument name="subject" default="Danielle of NewJeans"}, a mixed-race young woman with…

Brutalist Concrete Museum Interior

"Brutalist concrete museum interior, single sunbeam through skylight, sharp geometric shadows, monochromatic + one {argument name="accent color" default="red"} accent sculpture, {argument name="architectural style" default="Tadao Ando"} inspired"

Dreamy Balcony Guitar Portrait

Use the {argument name="reference" default="uploaded portrait"} only as the identity and hairstyle reference. Preserve the real face shape, facial features, skin tone, and hairstyle. Do not copy the original portrait’s clothes, background, pose, or lighting. Create a realistic dreamy soft-focus balcony photo of an {argument name="subject" default="adult young woman"} sitting by an old metal railing, gently playing an acoustic guitar. She wears a {argument name="clothing" default="loose white knitted cardigan, simple light inner top, casual shorts, and white socks"}. Natural relaxed pose with one leg crossed, head slightly lowered, eyes looking at the guitar, quiet and candid mood. Old apartment balcony, green leaves and vines outside, warm afternoon sunlight, dappled shadows on skin and clothes, soft haze, slight overexposure, film-like glow, real camera texture, natural skin, realistic hands and guitar details, no watermark, no text, no logo.

Master Model Maker Concept Template

Do this for {argument name="structure name" default="Eiffel Tower"}: Input Variable: [INSERT STRUCTURE] System Instruction: Act as a Cinematic Director. Create a hyper-realistic portrait of a Master Model Maker putting the final touches on a scale replica of the Input Structure. 1. Logic: Analyze the Input Structure to determine the Setting and the Character: IF Ancient/Fantasy (e.g., Colosseum, Castle): Room: A dusty, sunlit Renaissance Workshop (Stone & Wood). also showcasing "Previous Projects" (other models in the same genre). Character: An elderly Artisan in a leather apron. Materials: Plaster, Clay, Balsam Wood. IF Sci-Fi/Industrial (e.g., Death Star, spaceship): Room: A sterile, cold Industrial Design Lab (Concrete & Metal). also showcasing "Previous Projects" (other models in the same genre). Character: A younger Prop Master in a workshop jumpsuit or lab coat. Materials: Grey Resin (Kitbash), Styrene, Fiber Optics. IF Modern Architecture (e.g., Skyscraper): Room: A clean, white Architectural Firm (Glass & White Walls). also showcasing "Previous Projects" (other models in the same genre). Character: A minimalist Architect in a black turtleneck. Materials: White Foamcore, Acrylic, 3D Printed Plastic. 2. The Composition : The Subject: The Character is leaning over a massive, incredibly detailed Scale Model of the Input Structure. The Action: They are using Tweezers or a fine brush to place a specific "Micro-Detail" relevant to the lore. Death Star: Placing a tiny Turbo-Laser Turret. Colosseum: Dusting sand onto the arena floor. City: Gluing a tiny streetlamp. The Scale: Extreme scale contrast. The human hands look giant next to the tiny windows/bricks of the model. 3. The Environment : The Blueprint: On the table, beneath the tools, lies the correct reference document: Sci-Fi: Blue grid schematic or iPad. Fantasy: Yellowed parchment map. Modern: White vector floorplan. The Clutter: The tools scattered around must match the medium (e.g., Clay tools vs. Soldering Iron vs. X-Acto knives). 4. Lighting & Atmosphere: Lighting Match: The lighting must match the inferred room. Workshop: Warm God Rays (Dusty). Lab: Cool Fluorescent Tubes (Clean/Hazy). Focus: Shallow depth of field. Sharp focus on the tweezers and the contact point. Output: ONE image, 16:9 Aspect Ratio, Cinematic Photography, 8k Resolution.

Ink Sketchbook Zine Illustration

Transform any provided reference photo into a {argument name="aspect ratio" default="vertical 4:5"} stylized ink-drawn sketchbook illustration. Use the person, outfit, pose, accessories, pet, objects, mood, and overall character from the reference photo as the main subject. Do not copy fixed clothing or fixed phrases from previous examples. Adapt all visual details to the new reference image. Create a high-detail black ink and fineliner illustration on a white sketchbook page. The drawing should fill almost the entire frame, with no fabric element, no colored cloth, and no extra textile foreground. Composition: - one main full-body or main portrait illustration based on the reference photo; - 1-3 additional smaller sketch studies inspired by the same subject, such as side view, back view, close-up portrait, accessory detail, pet/object detail, or gesture pose; - small hand-drawn decorative elements around the subject: hearts, stars, sparkles, arrows, leaves, dots, mood symbols, tiny doodles; - balanced but dense sketchbook layout, editorial zine feeling. All handwritten text must be in {argument name="language" default="Russian"} only. The text should be random, natural, and relevant to the mood, outfit, season, activity, personality, or atmosphere of the reference photo. Do not repeat the same phrases every time. Art style: expressive handmade sketchbook aesthetic, nostalgic indie zine warmth, high-detail manga-influenced editorial line art, rich black India ink, fineliner, marker and brush-pen hatching, cross-hatching, varied pen nib strokes, subtle ink bleed, visible sketch construction lines. Render textures from the reference carefully: fabric folds, hair, accessories, shoes, bags, jewelry, pets, objects, skin shadows, material surfaces, and distinctive silhouette. Camera and format: vertical 4:5 format, close-up overhead flatlay photo of a white sketchbook page, slightly tilted top-down angle around 20 degrees, centered composition, drawing filling the screen, visible paper grain and sketchbook edge. Lighting: soft natural diffused {argument name="lighting" default="morning daylight"} from a window, even cool-key illumination, subtle diagonal page shadow, gentle softbox fill, neutral whites, deep blacks. Image quality: high-resolution print-quality output, 300 DPI, ultra-crisp ink detail, slight warm tone adjustment, contrast +15, clarity +10, subtle vignette, selective sharpening on ink strokes, minimal noise reduction to preserve paper texture. Final look: contemporary sketchbook photography meets editorial zine illustration, clean, polished, high-fidelity, suitable for print and portfolio presentation.