Felo

GPT Image 2Prompts

Updated every day

10,000+ free image and video prompts
142 days ago

Selfie to Studio Headshot

Selfie to Studio Headshot

This prompt turns a casual reference selfie into a polished professional studio headshot, useful for business profiles, LinkedIn photos, and corporate portraits.

Prompt
Using the provided reference image, transform the casual poolside selfie into a clean professional head-and-shoulders studio portrait. Keep the same person and the same polo shirt, but remove the outdoor resort background, other people, furniture, pool, and all vacation context. Reframe to a centered portrait crop with the subject facing the camera, shoulders visible, and the selfie angle corrected into a formal straight-on composition. Add a smooth dark teal studio backdrop with subtle vignetting and soft, flattering portrait lighting for a polished corporate profile-photo look.
AuthorMrDejie
142 days ago

Soccer Event Poster Redesign

Soccer Event Poster Redesign

A graphic design prompt instructing the AI to act as a premium designer to transform an existing poster into a bold new version.

Prompt
You are a premium graphic designer with a human feel. Transform this boring poster and make it a bold and attractive poster for an {argument name="event" default="upcoming soccer event"}.
142 days ago

Clean Japanese/Korean Style Photoshoot

Clean Japanese/Korean Style Photoshoot

A prompt for creating high-resolution, bright, and airy portrait photography in a clean 9-grid layout, focusing on natural light and minimalist style.

Prompt
Reference the facial features of the person in the image, generate a 9-grid photo collage, lifestyle candid shots, portrait photography, 85mm lens. Shallow depth of field, blurred background, high resolution, clear details. Wearing {argument name="服装" default="white shirt, white shorts"}, simple and clean outfit, indoor shooting, white background wall, near a window, natural light, soft sunlight, high-key lighting, bright and airy, soft shadows. {argument name="风格" default="Japanese fresh style, Korean portrait style"}, airy feel, clean picture, minimalist style, multi-pose shooting, standing, sitting, looking back, close-up, half-body, full-body.
AuthorAzyu
142 days ago

Fiery Burger King Whopper Poster

Fiery Burger King Whopper Poster

This prompt generates a dramatic Korean Burger King burger advertisement poster suited for fast-food marketing, social media promo art, or a realistic product campaign visual.

Prompt
A dramatic fast-food poster advertisement for Burger King, vertical composition, photorealistic studio food photography mixed with bold Korean promotional graphic design. Place the official Burger King logo centered at the very top. Below it, add huge distressed block Korean headline text in a stacked layout reading "신제품 와퍼 출시!" with "와퍼" in vivid deep red and the other words in textured off-white, all on a dark smoky black background. In the center, show one massive hero burger shot from a slightly low front angle, perfectly centered and filling most of the frame. The burger has 8 clearly visible layers and elements: a glossy sesame seed top bun, curly green leaf lettuce, thick red tomato slices, thick white onion rings, sliced pickles, streaks of ketchup, melted American cheese, a flame-grilled beef patty with heavy char marks, and a soft bottom bun. Add a tiny golden crown icon floating just above the top bun. The burger sits on a dark rustic wooden stump pedestal. Surround the lower half and side edges with dramatic orange flames, glowing embers, heat sparks, and smoky atmosphere to emphasize flame grilling. Near the bottom, place a secondary Korean slogan in bold text, centered, reading "불맛의 정점을 경험하라" with "불맛" in red and the remaining words in warm beige-gold, separated above by a small flame icon. At the very bottom, add a red badge-shaped label with a dark border containing the Korean text "한정 판매" in bold black. Use a high-contrast black, red, orange, and toasted brown color palette, intense cinematic lighting, glossy food highlights, crisp ingredient texture, premium commercial advertising finish, and a fiery indulgent mood.

Photoreal librarian presenting AI book

Photoreal librarian presenting AI book

This prompt generates a realistic commercial-style portrait of a librarian-like woman presenting an AI-themed book, suitable for promotional book ads or social media marketing visuals.

Prompt
A photorealistic vertical promotional portrait in a cozy modern library or bookstore interior, showing a young East Asian woman standing behind a wooden counter and presenting a large book toward the camera. She has long straight dark brown hair parted near the center, small delicate earrings, and a neat, friendly, professional appearance. She wears a white button-up shirt, a muted green striped necktie, and a beige apron tied at the waist with a stitched name patch reading “Librarian Ai-chan” and a small open-book icon. Her left hand holds the book upright at chest height, angled slightly toward the viewer, while her right hand is open beneath it in a classic product-introduction gesture. The book is the clear focal point: a glossy dark cover with dramatic orange title text reading “Pollo AI” twice, one large at the top and one glowing near the center over a cinematic landscape image, plus smaller Japanese subtitle and feature text and a circular P-style logo near the lower left. Place a transparent acrylic desk sign on the left side of the counter with Japanese text above and the large name “Ai-chan” below. On the right side of the counter, stack exactly 3 hardcover books horizontally, each with numbered spine labels and Japanese titles. In the softly blurred background, include a tall wooden bookshelf on the left with books, small potted plants, and a warm lamp; a framed wall poster on the upper right with Japanese writing; and a leafy indoor plant on the far right. Add a small white “Pollo.ai” watermark in the top-right corner. Use warm ambient lighting, shallow depth of field, soft beige and wood tones, realistic skin and fabric texture, clean commercial composition, and the polished look of a professional lifestyle advertisement for a book or AI-related publication. Keep the framing from mid-thigh to above the head in a portrait 3:4 aspect ratio.
142 days ago

World Civilization Spatiotemporal Map

World Civilization Spatiotemporal Map

A complex infographic prompt to generate a 5000-year history map of world civilizations across time and space.

Prompt
Visually display {argument name="Theme" default="5000-year spatiotemporal map of world civilizations"}, including both time and space dimensions, in a popular science style, with accurate details.
Authortski
142 days ago

Anime Sakura Movie Poster

Anime Sakura Movie Poster

A cinematic Japanese anime poster for a bittersweet girls' coming-of-age romance, ideal for film key visuals, teaser art, or dramatic promotional posters.

Prompt
A polished Japanese animated film poster in a soft, luminous springtime anime style. The main scene shows 2 teenage schoolgirls walking hand in hand down a quiet residential street lined with blooming cherry blossom trees. They wear matching Japanese sailor school uniforms: white long-sleeve blouses with navy sailor collars, pale blue neckerchiefs, and dark navy pleated skirts, each carrying a navy school bag. The girl on the left has dark hair tied in a low side bun with loose strands and a small pink cherry blossom hairpin; she turns toward the other girl with a tender, slightly wistful expression. The girl on the right has long straight dark hair and looks back at her softly. Pink sakura petals drift through warm late-afternoon backlight, creating a romantic, emotional atmosphere with glowing rim light, gentle lens bloom, and painterly detail. Along the left edge, add a vertical collage of 7 inset memory scenes from the story: 1) two children together in a nostalgic childhood moment, 2) a close portrait of the long-haired girl, 3) the two characters standing apart in a tense or reflective moment, 4) an indoor conversation scene with 3 people, 5) a close profile of the bun-haired girl, 6) an intimate almost-kiss close-up of the two leads, 7) a rear view of the two girls walking together beneath cherry blossoms. At the top, include the Japanese tagline: 「変わってしまった朝、変わらなかった想い。」 Across the lower center, place the large glowing handwritten title {argument name="title text" default="Story of U"} in elegant pink neon script. Beneath it, add small Japanese subtitle text and a release line reading {argument name="release text" default="2027年春公開"}. At the very bottom, add dense small white Japanese credits in authentic movie-poster layout. Overall mood: bittersweet yuri coming-of-age drama, cinematic key visual, emotionally resonant, highly detailed, pastel pink and navy palette, vertical poster composition.
142 days ago

Chinese Alzheimer’s Infographic Poster

Chinese Alzheimer’s Infographic Poster

A detailed traditional-style Chinese medical infographic poster explaining Alzheimer’s disease for health education, awareness campaigns, and social media science posts.

Prompt
{"type":"traditional Chinese medical infographic poster","topic":"{argument name=\"topic\" default=\"Alzheimer's Disease\"}","title":"{argument name=\"headline text\" default=\"Alzheimer's Disease - Detailed Infographic\"}","language":"English","style":{"overall":"educational health infographic in vintage Chinese poster style","rendering":"clean hand-painted illustration, soft watercolor textures, thin ink outlines, muted pastel colors","palette":"teal, beige, cream, jade green, dusty blue, warm brown, light coral","decorative frame":"ornate scroll-like border with gold trim","background":"pale sky-blue backdrop with traditional Chinese landscape motifs, stylized clouds, pine trees, mountains, cranes, and waves"},"layout":{"format":"vertical poster","sections":[{"title":"1. Disease Overview","position":"top-left","count":1,"labels":["Cognitive Decline","Memory Loss"],"content":"one large panel explaining that Alzheimer disease is a progressive neurodegenerative disorder, featuring an elderly side-profile portrait with a semi-transparent brain illustration showing degeneration and small neuron icons around the brain"},{"title":"2. Common Symptoms","position":"top-right","count":6,"labels":["Mild","Moderate","Severe","Forgetting","Emotional Fluctuation","Speech Difficulty","Disorientation","Loss of Self-care"],"content":"a symptom panel arranged as 3 columns by severity, with 6 vignette illustrations of elderly people: forgetting at a table, emotional fluctuation, headache or confusion, speech difficulty, getting lost outdoors, and losing ability for daily self-care with assistance from another person"},{"title":"3. Pathological Causes","position":"middle-left","count":4,"labels":["Amyloid Plaques","Aβ deposits","Tau Tangles","Neurofibrillary tangles","Neuron Loss","Apoptosis"],"content":"a cross-section anatomical brain diagram with bilingual annotations pointing to amyloid beta deposits, tau protein tangles, neurofibrillary tangles, and neuron loss"},{"title":"4. Risk Factors","position":"middle-right","count":5,"labels":["Age","Genetics","Cardiovascular Disease","Education Level","Head Trauma"],"content":"five icon-based risk factors shown in a grid: an elderly bust for age, DNA helix for genetics, anatomical heart for cardiovascular disease, stacked books for education level, and a head profile with bandage for head trauma"},{"title":"5. Diagnostic Methods","position":"bottom-left","count":4,"labels":["Cognitive Test","MMSE","Brain Imaging","Biomarkers"],"content":"a diagnosis panel with a doctor consulting an elderly patient at a desk, plus three diagnostic visuals: an MMSE test sheet, a brain imaging scan with MRI-like machine, and a test tube representing biomarkers"},{"title":"6. Treatment & Care","position":"bottom-right","count":7,"labels":["Medication","Pills","Donepezil","Memantine","Non-drug Therapy","Elderly doing Tai Chi","Music","Painting","Care & Support","Family Support","Understanding, Patience, Respect"],"content":"a care and treatment panel divided into 3 columns: medication with two pills and the words Donepezil and Memantine, non-drug therapy with tai chi, music, and painting, and caregiving support with family companionship plus a smaller scene emphasizing understanding, patience, and respect"}],"header":{"count":1,"text":"large centered Chinese calligraphy title banner at the top"},"panel_style":"six rounded rectangular cream panels with decorative teal-and-gold title tabs"},"composition":{"symmetry":"balanced two-column arrangement with three rows of paired panels","readability":"clear educational hierarchy with large title, numbered section headers, medium illustrations, and bold labels","density":"information-dense but visually friendly"},"negative_constraints":["no photorealism","no modern flat UI","no dark horror tone","no clutter outside the six main panels","no extra sections beyond the 6 numbered panels"],"quality":"high-resolution, polished infographic suitable for public health education, hospital waiting rooms, science communication posts, or social media comparison graphics"}
142 days ago

Chinese Alzheimer’s Awareness Infographic

Chinese Alzheimer’s Awareness Infographic

A traditional Chinese-style medical infographic poster explaining Alzheimer’s disease, designed for public health education, awareness campaigns, and social media comparison posts.

Prompt
{"type": "traditional Chinese medical infographic poster", "topic": "Alzheimer's Disease", "style": {"overall": "elegant health education poster in Chinese ink-wash and vintage parchment style", "palette": "warm beige background, dark brown text, muted red title banners, soft gray ink landscape accents", "rendering": "clean infographic mixed with hand-painted watercolor illustration and light line-art icons", "mood": "serious, informative, compassionate"}, "canvas": {"orientation": "portrait", "composition": "single-page poster with header, multiple boxed information panels, and footer slogan"}, "text": {"main_title": "Alzheimer's Disease", "subtitle": "—— Understand, Prevent, Care, Build the Great Wall of Memory Together ——", "body_language": "Simplified Chinese", "footer_slogan": "Early detection, early diagnosis, early intervention. Caring for memory starts with you and me!", "footer_note": "If you notice related symptoms in people around you, please seek medical attention promptly."}, "decorative_elements": {"count": 6, "items": ["plum blossom branch across the upper right corner", "misty mountain landscape in the background", "lotus flowers and leaves at lower left", "small boat on water near left-center bottom", "bamboo illustration beside the diagnosis section", "distant mountains and pavilion at lower right with flying birds"]}, "layout": {"sections": [{"title": "What is Alzheimer's Disease?", "position": "upper left", "count": 4, "labels": ["A chronic disease with insidious onset", "Mainly manifested by memory loss", "As the disease progresses, it affects language, thinking, behavior, emotions, and daily living abilities", "There is currently no cure, but early intervention can delay progression and improve quality of life"]}, {"title": "Disease Mechanism (Brain Changes)", "position": "upper right", "count": 4, "labels": ["Abnormal protein deposits in the brain lead to neuronal damage and death, causing gradual cognitive decline.", "Normal brain", "Alzheimer's patient's brain", "Amyloid-beta plaques and Tau protein neurofibrillary tangles"]}, {"title": "Risk Factors", "position": "middle right", "count": 5, "labels": ["Aging (significant risk increase >65 years)", "Genetic factors (e.g., APOE ε4 gene)", "Cardiovascular diseases (hypertension, diabetes, etc.)", "Lower education level, insufficient cognitive activity", "Lack of exercise, smoking, obesity, depression, etc."]}, {"title": "Clinical Manifestations (Disease Progression Stages)", "position": "middle to lower center", "count": 4, "labels": ["Early stage (Mild Cognitive Impairment)", "Middle stage (Mild Dementia)", "Middle-late stage (Moderate Dementia)", "Late stage (Severe Dementia)"]}, {"title": "How to Diagnose?", "position": "lower left", "count": 4, "labels": ["Detailed medical history collection", "Cognitive function assessment", "Imaging examination", "Laboratory tests"]}, {"title": "Prevention and Delay Suggestions (Healthy Life Starts Now)", "position": "lower right", "count": 7, "labels": ["Healthy diet: Mediterranean diet, eat more fruits, vegetables, whole grains, fish, nuts.", "Regular exercise: At least 150 minutes of moderate-intensity exercise per week.", "Active brain: Reading, learning new skills, playing chess, calligraphy, etc.", "Social activities: Maintain interpersonal relationships, actively participate in socializing.", "Control chronic diseases: Manage blood pressure, blood sugar, blood lipids, quit smoking and limit alcohol.", "Good sleep: Ensure adequate sleep, treat sleep apnea.", "Stay optimistic: Positive attitude, reduce stress and depression."]}], "centerpiece": "a large multi-panel educational layout with red section headers and thin ornamental borders"}, "illustrations": {"count": 14, "items": ["2 cross-sectional brain diagrams comparing healthy and Alzheimer’s brains", "2 circular pathology insets showing amyloid plaque and tau tangle", "5 round risk-factor icons", "4 patient-stage illustrations showing progression from mild confusion to bedridden late stage", "1 clipboard icon in diagnosis section plus supporting medical-style icons"]}, "typography": {"headline": "large expressive black brush-calligraphy title", "section_headers": "white Chinese text on dark red brushstroke banners", "body": "neat printed Chinese serif text with moderate spacing"}, "quality": "high-resolution, balanced editorial composition, polished healthcare infographic suitable for social media sharing, public education, or clinic awareness poster"}
142 days ago

Back-to-Back Sirens in Sakura Dusk

Back-to-Back Sirens in Sakura Dusk

A cinematic fantasy artwork of two tattooed female warriors standing back-to-back in a cherry blossom field at sunset with two glowing energy swords, ideal for dramatic character art and poster-style hero imagery.

Prompt
Two powerful fantasy women standing back-to-back in the center of a vast cherry blossom field at dusk, full-body heroic composition, seen from a slightly low angle, ultra-detailed cinematic anime-realism. The woman on the left has {argument name="left character hair color" default="vivid crimson red"} short layered hair, a curvy athletic build, and a fierce dominant presence; she wears a black and dark red strappy battle outfit with a cropped leather top, high-cut bottoms, garter-like thigh straps, flowing translucent red ribbons, and long glossy black armguards. The woman on the right has {argument name="right character hair color" default="golden blonde with soft pink ends"} wavy shoulder-length hair, an athletic curvy build, and extensive floral tattoo sleeves and body tattoos across her arm, shoulder, ribs, and thigh; she wears a black and purple strappy fantasy outfit with sheer purple fabric panels, chain accents, and black gloves. Their bare upper backs press together as they face opposite directions with calm, intimidating confidence. There are exactly 2 massive glowing energy swords planted upright in the ground behind and beside them: 1 on the left radiating molten red-orange fire and ember-like cracks, and 1 on the right radiating electric violet and cyan energy with lightning coursing through the blade. The scene is set in a breathtaking sakura landscape with dense pink cherry trees, countless petals swirling violently through the air, a dramatic sunset sky filled with magenta, purple, and orange clouds, distant lightning, and a glowing horizon over a Japanese-inspired cityscape with a pagoda visible on the right. Emphasize raw power, sensual elegance, high contrast rim lighting, wind-blown fabric and hair, luminous particle effects, rich red-versus-purple color contrast, epic fantasy atmosphere, hyper-detailed rendering, sharp focus on the two women, and painterly cinematic depth.
142 days ago

Emerald Dark Sorceress Ritual

Emerald Dark Sorceress Ritual

This prompt generates a dramatic dark-fantasy character illustration of a silhouetted sorceress before a glowing arcane circle, ideal for gothic magic-themed key art or book-cover-style imagery.

Prompt
A dark fantasy full-body portrait of a mysterious sorceress standing barefoot in the center of an ancient ritual chamber, shown in side profile with her head slightly bowed and her body mostly in silhouette against an intense glowing emerald magic circle. She has {argument name="hair style" default="dark curly hair in a loose messy bun with long tendrils framing her face"} and wears a {argument name="dress color" default="black"} floor-length gown with a very low open back, a thigh-high slit, and long torn shadow-like ribbons or vine-like wisps trailing from the dress and wrapping around her body. Behind her is 1 enormous circular arcane sigil composed of many concentric rings, geometric lines, occult symbols, rune-like markings, and 3 prominent pentagram medallions positioned at the top, left, and right. The environment is a gothic stone hall with tall shadowy pillars, drifting green smoke, faint magical haze, and a glossy reflective floor engraved with 1 additional partial summoning circle in the foreground. Use a monochrome palette of {argument name="glow color" default="neon emerald green"} and black, with extreme contrast, dramatic backlighting, volumetric fog, cinematic rim light, intricate magical linework, ethereal smoke, and a haunting supernatural atmosphere. Make the figure elegant, dangerous, and otherworldly, with ultra-detailed fantasy illustration rendering, symmetrical composition centered on the glowing sigil, vertical aspect ratio, and a mood of forbidden ritual magic.
142 days ago

Crimson Dark Sorceress Ritual

Crimson Dark Sorceress Ritual

This prompt generates a dramatic dark-fantasy character poster of a floating sorceress silhouetted against a glowing red magic circle, ideal for gothic concept art or cinematic cover imagery.

Prompt
A dark fantasy illustration of a single floating woman centered in the frame, shown as a near-black silhouette against an intense glowing crimson background. She hovers a little above a reflective black-red water surface, barefoot, with both arms relaxed and slightly extended downward, fingers softly spread, her posture calm but powerful. Her very long {argument name="hair color" default="black"} hair streams dramatically to the left in wild windblown tendrils, and her gown is a shredded, smoky, semi-translucent {argument name="dress color" default="black"} dress with layered ribbons and torn fabric trailing downward into shadowy wisps. Behind her is one enormous ornate circular magic sigil, perfectly framing her body like a halo, rendered in intricate gothic filigree, concentric rings, and arcane lace-like geometry, glowing bright blood red with a hot white-red core. The entire scene is dominated by a black and deep red palette, with swirling smoke, embers, and magical energy filling the air. At the lower left and lower right edges are 2 clusters of sharp red crystals emerging from darkness. The bottom of the image shows one glossy reflective pool with rippling crimson highlights mirroring the light above. Make the atmosphere infernal, mystical, and regal, as if {argument name="character role" default="a dark sorceress"} is awakening in a forbidden ritual chamber. High contrast, cinematic backlighting, volumetric glow, ultra-detailed digital painting, vertical composition, centered symmetrical framing, ominous elegance, no text, no extra characters, no visible facial details.
142 days ago

Full Stack Knowledge Visualization

Full Stack Knowledge Visualization

A prompt that creates a visual representation of full-stack development knowledge for VibeCoding, including top project demos.

Prompt
Visual representation of {argument name="Subject" default="basic Web full-stack knowledge needed by ordinary people for VibeCoding"}, and {argument name="Quantity" default="Top 10"} full-stack demos that can be built.
142 days ago

DeepSeek V3 vs V4 Architecture Infographic

DeepSeek V3 vs V4 Architecture Infographic

A dense side-by-side technical infographic comparing DeepSeek V3/R1 and DeepSeek V4 transformer architectures, suitable for social media posts, presentations, or model analysis visuals.

Prompt
{"type":"side-by-side AI architecture comparison infographic","style":"clean technical diagram, white background, thin black outlines, rounded rectangles, dashed callout boxes, color-coded highlights, presentation-slide aesthetic, vector infographic","canvas":{"aspect_ratio":"2:1","resolution":"wide horizontal"},"title_row":{"left_title":"DeepSeek V3/R1 (671 billion)","right_title":"DeepSeek V4 (1.2 trillion)","left_title_color":"bright orange-red","right_title_color":"bright blue"},"layout":{"columns":2,"sections":[{"title":"DeepSeek V3/R1 (671 billion)","position":"left half","count":9,"labels":["Vocabulary size of 129k","FeedForward (SwiGLU) module","Intermediate hidden layer dimension of 2,048","MoE layer","Supported context length of 128k tokens","First 3 blocks use dense FFN with hidden size 18,432 instead of MoE","Sample input text","Embedding dimension of 7,168","128 heads"]},{"title":"DeepSeek V4 (1.2 trillion)","position":"right half","count":9,"labels":["Vocabulary size of 160k","FeedForward (SwiGLU) module","Intermediate hidden layer dimension of 3,072","MoE layer","Supported context length of 256k tokens","First 3 blocks use dense FFN with hidden size 24,576 instead of MoE","Sample input text","Embedding dimension of 8,192","128 heads"]},{"title":"bottom comparison table","position":"bottom full width","count":10,"labels":["Total parameters","Active parameters per token","Hidden size","Esmple dimesiegn","DeepSeek V3/R1","Intermediate (FF)","Attention heads","Context length","Embedding dimension","Vocabulary size"]}]},"left_panel":{"background":"very light gray rounded rectangle","main_stack":{"count":8,"blocks":["Tokenized text","Token embedding layer","RMSNorm 1","Multi-head Latent Attention","RMSNorm 2","MoE","Final RMSNorm","Linear output layer"]},"side_module":"RoPE attached to the attention block on the left side","attention_block":{"label":"Multi-head Latent Attention","accent":"orange-red text for the word Latent"},"feedforward_inset":{"title":"FeedForward (SwiGLU) module","count":4,"blocks":["Linear layer","SiLU activation","Linear layer","Linear layer"],"diagram":"two branches multiplied, then projected"},"moe_inset":{"title":"MoE layer","count":5,"blocks":["top combine node","Feed forward","Feed forward","Router","expert count badge 256"],"details":"small black square with 1 selected expert, arrows routing upward to experts, dotted divider line"},"annotations":{"vocab":"Vocabulary size of 129k","ff_dim":"Intermediate hidden layer dimension of 2,048","context":"Supported context length of 128k tokens","dense_first_blocks":"First 3 blocks use dense FFN with hidden size 18,432 instead of MoE","resource_savings":"Resource savings: Model size is 671B but only 1 (shared) + 8 experts active per token; only 37B parameters are active per inference step"},"bottom_stats":{"count":10,"items":["Total parameters: 671B","Active parameters per token: 37B (1 + 8 experts)","Hidden size: 7,128","Esmple dimesiegn: 28,432","Intermediate (FF): 2,048","Attention heads: 128","Context length: 128k","Embedding dimension: First 3 blocks","Context ler length: 22G7","Vocabulary size: 129k"]}},"right_panel":{"background":"very light blue rounded rectangle","main_stack":{"count":8,"blocks":["Tokenized text","Token embedding layer","RMSNorm 1","Multi-head Latent Attention","RMSNorm 2","MoE","Final RMSNorm","Linear output layer"]},"side_module":"RoPE attached to the attention block on the left side","attention_block":{"label":"Multi-head Latent Attention","accent":"blue text for the word Latent"},"feedforward_inset":{"title":"FeedForward (SwiGLU) module","count":4,"blocks":["Linear layer","SiLU activation","Linear layer","Linear layer"],"diagram":"same structure as left panel"},"moe_inset":{"title":"MoE layer","count":5,"blocks":["top combine node","Feed forward","Feed forward","Router","expert count badge 384"],"details":"small black square with 1 selected expert, arrows routing upward to experts, dotted divider line, blue border emphasis"},"annotations":{"vocab":"Vocabulary size of 160k","ff_dim":"Intermediate hidden layer dimension of 3,072","context":"Supported context length of 256k tokens","dense_first_blocks":"First 3 blocks use dense FFN with hidden size 24,576 instead of MoE","resource_savings":"Resource savings: Model size is 1.2T but only 1 (shared) + 8 experts active per token; only 52B parameters are active per inference step"},"bottom_stats":{"count":10,"items":["Total parameters: 1.2T","Active parameters per token: 52B (1 + 8 experts)","Hidden size: 7,2B","Esmple dimesiegn: 28,432","Intermediate (FF): 3,072","Attention heads: 128","Context length: 256k","Embedding dimension: First 3 blocks","Context ler length: 22G7","Vocabulary size: 160k"]}},"global_notes":"Create a highly detailed transformer architecture comparison diagram with mirrored layouts. Each half contains one large model stack diagram plus 2 inset diagrams: 1 feedforward module and 1 MoE layer. Use arrows between blocks, tiny technical labels, and connector lines from labels to the relevant components. Keep the typography dense and slide-like, with orange-red used for all V3/R1 emphasis and blue used for all V4 emphasis. Include a small bottom row of compact tabular metrics spanning the width. Preserve the slightly imperfect, human-made infographic look with very small text and crowded annotations."}
142 days ago

Demon AI Intern Office Poster

Demon AI Intern Office Poster

This prompt generates a stylized vertical Chinese promo poster of a horned red-haired AI office assistant in a warm startup workspace, ideal for character branding, meme ads, and social-media marketing.

Prompt
Create a dramatic Chinese social-media poster in a warm amber office setting, featuring a stylish demon-themed female assistant character seated at a desk. The composition is a single vertical poster with dense graphic design, cinematic lighting, and a premium glossy illustration style mixing anime realism and commercial advertising. At the top, place a huge bold red Chinese headline reading {argument name="headline text" default="龙虾夏夏"}, with an elegant handwritten script overlay reading {argument name="signature text" default="Xia Xia"}. Under it, add a strong black Chinese slogan: {argument name="subtitle text" default="嘴欠,但能交付"}. The character is centered, upper body visible behind a desk, with messy vivid red hair in a loose bun and many flyaway strands, small dark red devil horns, pale skin, black layered clothing, a lanyard ID badge, long red metallic claw gloves, and a tired but dangerous office-worker aura. Her face is intentionally obscured by a square censor block in the center. She props her cheek with one clawed hand and holds a branded black takeaway coffee cup in the other hand. Add subtle lobster-inspired details in the outfit and props, including segmented glossy red shell-like forms behind and around her chair. The desk contains exactly 7 visible items: 1 open laptop with a red lobster logo and the words "XIA XIA", 1 closed dark book stack beneath the laptop, 1 open notebook with handwritten sketches and Chinese notes, 1 black fountain pen, 1 rectangular black desk nameplate with Chinese text, 1 takeaway coffee cup with matching lobster logo, and 1 hanging employee badge card. In the left background, show a computer or presentation screen displaying a Feishu-style document interface with the Chinese title "内容创作流程" and a simple 3-step flowchart labeled exactly "场景", "流程", "资产", plus the sentence "Agent 是放大器,不是方向盘。" Add a sticky note near that screen reading "扣瑞芋". In the upper left, include a yellow annotation box with the Chinese text "黄叔雇来的毒舌但能干的AI实习生" and a playful doodled face beside it. In the right background, pin up 3 office notes: 1 clipboard-style intern profile card titled "实习生档案" with multiple bullet lines of Chinese profile text and a handwritten sign-off, 1 checklist board titled "今日待办" with 5 checkbox items and a red star on the final line, and 1 yellow sticky note reading "老板别演". Make all Chinese typography feel intentionally designed, as if for a meme ad campaign and character branding poster. At the bottom, add a bold centered Chinese footer slogan reading {argument name="footer slogan" default="不演龙虾,我就是龙虾。"}. Use orange sunset backlight, warm shadows, glossy reflections on the claws, a cluttered but stylish startup office atmosphere, black and burnt-orange color grading, and a humorous yet competent "sharp-tongued AI intern" vibe. Include visible branding text "XIA XIA" on the cup and laptop, and make the whole image feel like a high-end character promo poster for an assistant persona used in workplace AI marketing.
AuthorJohnn
142 days ago

Noir Woman With Red Halo Poster

Noir Woman With Red Halo Poster

This prompt generates a minimalist noir-inspired art poster of a woman in profile with dripping ink clothing and a bold red circle, ideal for editorial-style wall art or dramatic cover imagery.

Prompt
A minimalist fine-art poster portrait of a woman in clean side profile facing right, shown from about mid-torso up, with her face partially obscured by a soft rectangular blur over the front of the head. She has dark black hair styled in a low loose bun with gentle volume and wispy strands, and wears 1 dangling silver earring on the visible ear. Dress her in a structured black blazer or tailored coat with a deep neckline. The lower half of the figure dissolves into elongated dripping black ink trails, with many uneven vertical streaks extending downward like wet paint pulled by gravity, creating a dramatic melting silhouette. Place 1 large, perfectly round vivid red disc directly behind her head and upper shoulders, centered like a halo or rising sun. Use a strictly limited palette of black, red, and warm off-white. Set the background to a soft cream paper or canvas texture with subtle grain. Light the subject like a high-end noir editorial portrait with strong contrast, sculpted highlights on the ear, jaw, neck, and cheek contour, and deep shadow across the clothing. Keep the composition tall and centered with generous negative space, elegant poster-like balance, moody contemplative atmosphere, and crisp, refined photographic detail blended with painterly graphic design.
AuthorJohnn
142 days ago

Noir Man With Red Sun Poster

Noir Man With Red Sun Poster

A minimalist surreal poster of a suited man in side profile dissolving into ink drips before a red circle, ideal for editorial art, gallery-style prints, or cinematic noir branding.

Prompt
Create a minimalist vertical art poster featuring a single man in clean side profile facing left, shown from about mid-thigh upward, wearing a tailored dark suit jacket, crisp white shirt, and a vintage 1940s-style black fedora with a ribbon band. His pose is upright, composed, and confident, evoking a classic noir protagonist. Render the figure in a refined monochrome photographic style with crisp editorial detail, rich tonal contrast, and dramatic Rembrandt-inspired lighting that sculpts the cheek, ear, jawline, hat brim, and shoulder with deep shadow. Place 1 bold perfectly round red disc directly behind the head and upper torso, centered like a halo or rising sun, creating a stark graphic focal point. The background should be a clean off-white or cream paper-textured surface with subtle grain and generous negative space around the subject. The lower half of the figure should dissolve into elongated black ink drips: count 12 distinct major dripping trails of varying thickness, hanging from the jacket hem and lower body, some long and straight, some tapering, with a few small splatters, as if the body is bleeding into the canvas. Keep the color palette strictly limited to black, red, and warm off-white. The overall mood should be moody, contemplative, elegant, slightly ominous, and modernist, like a high-end gallery poster blending film noir portraiture with surreal ink-paint abstraction.
142 days ago

Book Visualization Prompt

Book Visualization Prompt

A prompt for GPT Image 2 to visualize core concepts from 'The Almanack of Naval Ravikant' into shareable social media content.

Prompt
Visualize the core content of {argument name="Book Title" default="The Almanack of Naval Ravikant"}, use specific scene illustrations to make it concrete, and include elements of viral spread.