Промпты для нейросетей
Чат-бот: @aicreator Сообщество → @creatory Делимся фишками и лайфхаками различных нейросетей. Обучаемся, обмениваемся опытом. Написать нам: @let_adm
Show more📈 Analytical overview of Telegram channel Промпты для нейросетей
Channel Промпты для нейросетей (@prompty) in the Russian language segment is an active participant. Currently, the community unites 29 975 subscribers, ranking 939 in the Art & Design category and 21 359 in the Russia region.
📊 Audience metrics and dynamics
Since its creation on невідомо, the project has demonstrated rapid growth, gathering an audience of 29 975 subscribers.
According to the latest data from 06 November, 2026, the channel demonstrates stable activity. Although there has been a change in the number of participants by -78 over the last 30 days and by 6 over the last 24 hours, overall reach remains high.
- Verification status: Not verified
- Engagement rate (ER): The average audience engagement rate is 10.38%. Within the first 24 hours after publication, content typically collects 3.81% reactions from the total number of subscribers.
- Post reach: On average, each post receives 3 111 views. Within the first day, a publication typically gains 1 142 views.
- Reactions and interaction: The audience actively supports content: the average number of reactions per post is 20.
- Thematic interests: Content is focused on key topics such as nanobanana, внешность, soft, clean, background.
📝 Description and content policy
The author describes the resource as a platform for expressing subjective opinions:
“Чат-бот: @aicreator
Сообщество → @creatory
Делимся фишками и лайфхаками различных нейросетей.
Обучаемся, обмениваемся опытом.
Написать нам: @let_adm”
Thanks to the high frequency of updates (latest data received on 07 November, 2026), the channel maintains relevance and a high level of publication reach. Analytics show that the audience actively interacts with content, making it an important point of influence in the Art & Design category.
Перерисуйте фотографию в виде уютной скандинавской иллюстрации для поздравительной открытки, сохранив людей, позы и одежду. Цифровая гуашь сухой кистью на обычной белоснежной акварельной бумаге, пустой фон. Удлиненные фигуры, маленькие головы, объемная одежда, тонкие ноги, минималистичные лица с точечными глазами и розовыми щеками. Приглушенный зеленый, бежевый, коричневый, серый, один кораллово-красный акцент, маленькое красное сердечко вверху.💬 Обсудить в @creatory ⭐️ Повторить в @aicreator #GPTimage25 😀😀😀😀😀
Преврати фото в авторскую наивную рождественскую иллюстрацию: hand-drawn, минимализм, забавные пропорции, неровные контуры, плоские контрастные цвета, винтажная фактура бумаги, декоративные штрихи. Без теней, градиентов и 3D. Эстетика дорогой дизайнерской открытки, уютная праздничная атмосфера. Сохрани узнаваемость человека и основные детали фото. Формат А6, вертикальный.💬 Обсудить в @creatory ⭐️ Повторить в @aicreator #GPTimage25 😀😀😀😀😀
Use my uploaded photo as the face reference. 3:4 vertical. Point-of-view shot from inside a grey fleece-lined bag looking up. The soft fuzzy lining frames every edge of the image. the person in reference, keeping his exact face, hairstyle and identity, reaches down into the bag toward the camera, his fingers foreshortened in the foreground and stacked with chunky silver rings. He wears a black ribbed beanie, black oval sunglasses, a navy blazer and a blue-and-white striped shirt with silver cross collar pins. Behind him: a white London townhouse with a dark grey-green front door numbered 27, black iron railings and a small olive tree. Overlaid iOS incoming call interface in white system font: a "Message" button on the left and a "Voicemail" button on the right as frosted round icons, and at the bottom a frosted glass pill slider reading "slide to answer" with a white circle holding a green phone icon. Natural daylight, candid street-style fashion photography, realistic skin texture.Взгляд из стаканчика:
Use the uploaded photo for the face with the prompt below. Preserve the person's features exactly: appearance, age, hair, eyes, skin tone and texture — identical to the original. Shoot from deep inside a [Coctail] cup, aiming straight up. Use an extreme fisheye lens to barrel-distort the entire frame. The foreground is a wall of crushed ice in macro detail, glistening, textured, and richly colored, lit with bokeh highlights. A single straw cuts a perfect vertical line through the entire composition, from the person's pursed lips at the top all the way down into the ice bed below. The person peers down through the cup opening, centered in the circular frame of the sky, wide-eyed, thrilled, lips firmly around the straw. The cup wall rings the entire scene with its branded logo and colors. Above them: brilliant clear blue sky. Colors are punchy and complementary, true to the drink's natural palette. Hyper-real commercial photography quality, blazing summer daylight.Рыбий глаз:
Use the uploaded photo for the face with the prompt below. Preserve the person's features exactly: appearance, age, hair, eyes, skin tone and texture — identical to the original. Ultra-realistic action photography, extreme fisheye lens, 8mm, worm’s-eye view from ground level, dynamic urban plaza scene. A stylish young person appears to be stepping over the camera, one sneaker sole dominating the foreground, hand reaching toward the lens.. Massive circular “tiny planet” perspective with historic European architecture curving around the edges of the frame, bright blue sky filled with fluffy white clouds at the center. Cinematic composition, dramatic distortion, ultra-wide angle, vibrant colors, sharp focus, HDR, photorealistic, highly detailed skin texture, travel photography aesthetic, motion energy, depth and scale, 8K resolution, professional DSLR quality, masterpiece, Instagram viral shot. Aspect Ratio 9:162️⃣Генерация видео: Заходите в ChatGPT или любую другую языковую модель, загружаете полученное фото и просите придумать промпт под видео. ✈️Далее заходите в @aicreator 👉 😀😀😀😀 👉 ⭐️ 👉 выбираете «Seedance 2.5» 👉 вставляете полученное фото и промпт. 🔖Свои результаты показывайте в комментариях ниже, будем рады посмотреть все ваши работы!
Transform the uploaded photograph into a minimalist handmade felt applique illustration. Imagine the photograph has been simplified into a small piece of Korean handmade fabric artwork. Use soft felt and wool textures. Simplify all elements into large rounded shapes. The person should be recognizable through: silhouette, hairstyle, pose, clothing, and a few characteristic details. Keep facial features extremely minimal. Use soft muted colors: cream, milk beige, oatmeal, dusty pink, sage green, soft brown, pale blue. Create subtle fuzzy felt edges and handmade imperfections. The artwork itself should be very small within a large warm-white fabric background. Extremely generous negative space. Soft. Cute. Warm. Tactile. Minimal. It should look like a tiny handmade Korean lifestyle goods artwork. NO text. NO realistic face. NO photorealism. NO dense details.💬 Обсудить в @creatory ⭐️ Повторить в @aicreator #GPTimage25 😀😀😀😀😀
請把上傳照片當作不可重畫的鎖定底圖,進行「主體輪廓亂碼貼紙 × 檔案印刷故障」風格編輯。保留原圖主體數量、位置、大小、姿態、構圖和背景關係,不裁切、不擴圖、不重構。沿著人物或主要物體真實輪廓加入暖白色紙張貼片,內部填充無意義的黑色等寬字母、數字和符號;疊加少量黑色點陣、半色調、棋盤格、像素邊緣和輕微錯版效果,同時保留約一半原圖色彩和紋理。整體加入舊雜誌印刷、影印顆粒、紙張纖維和輕微掉色質感,色彩略降純度,從原圖提取一種亮色作為少量視覺點綴。效果要像手工拼貼雜誌,不要均勻濾鏡,不改變人物臉、手、姿態和原本場景。💬 Обсудить в @creatory ⭐️ Повторить в @aicreator #GPTimage25 😀😀😀😀😀
Cinematic dramatic portrait of person from the uploaded image glowing red from side light, head bowed down in shadow, wearing black jacket and white shirt, vibrant orange and teal bokeh light streaks in dark background, high contrast rim lighting, moody atmospheric film photography style💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
Создай фотореалистичный вертикальный портрет, используя лицо человека с загруженного селфи. Сохрани его узнаваемые черты лица, форму носа, губ, подбородка, пропорции и естественную текстуру кожи. Человек лежит на спине, его голова окружена плотным слоем крупных спелых оранжевых апельсинов, заполняющих весь кадр. Вид строго сверху, лицо расположено по центру. На человеке яркие бирюзовые солнцезащитные очки с цветочным узором на оправе и тёмными линзами. Волосы слегка растрёпаны и разложены между помидорами. Во рту зажата тонкая сигарета, направленная по диагонали влево. Крупный план, симметричная композиция, насыщенные красные и оранжевые оттенки, тёплый направленный свет, глубокие естественные тени, реалистичные блики на кожице помидоров, высокая детализация, лёгкая плёночная фактура, эстетика необычной модной фотосъёмки. Лицо полностью видно, резкость на лице и очках. Без текста, логотипов, лишних предметов и дополнительных людей.💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
Используй фото 1 как строгий референс внешности человека: сохрани черты лица, возраст, оттенок кожи, цвет глаз, цвет и длину волос. Фотореалистичный портрет в тёмном кинозале, вертикальный формат. Человек сидит по центру в красном кресле, расслабленно откинувшись, спокойно смотрит прямо в камеру и пьёт молочный коктейль через трубочку из светлого стакана с красным рисунком. Другая рука погружена в огромное красно-белое ведро попкорна, занимающее нижний передний план. Позади слева — жуткий клоун с вытянутой белой маской и открытым овальным ртом, справа — Пеннивайз с рыжими волосами и зловещей улыбкой. Ещё один клоун частично обрезан справа на переднем плане. Светлые костюмы с пышными воротниками и красными помпонами. Клоуны не перекрывают лицо человека. Над головой — крупный красный воздушный шар, дополнительные красные шары по краям. Приглушённая прямая вспышка, тёмный фон, слегка зеленоватые тени, тёплый свет на коже, плёночное зерно и лёгкая дымка. Лицо в фокусе, ближайший попкорн немного размыт. Эстетика винтажного Halloween-снимка на компактную камеру.💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
A hyper-realistic cinematic photograph of the person from the reference image standing in a field, extreme low angle worm's-eye view shot looking up at them. Preserve the person's face, facial features, hairstyle and hair color exactly as in the reference, with hair slightly tousled by the wind. The person is wearing the exact same outfit as in the reference image — same clothing items, colors, patterns, fabric and fit — with natural fabric folds and realistic draping, moving slightly in the breeze. They are looking slightly down at the camera with a calm, gentle, and nostalgic gaze. The immediate foreground features sharp, tall lush green grass blades framing the bottom. The background is a vast, vibrant azure blue sky with fluffy white cumulus clouds. Warm golden hour sunlight, subtle sun flare from the right, cinematic rim lighting highlighting the hair and shoulders. Vivid saturated colors, nostalgic summer vibe, shot on 35mm portrait lens, photorealistic, 8k resolution, highly detailed natural skin texture, slight film grain, cinematic color grading💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
Use the attached image as the exact base image. Preserve its original colours, subject, pose, composition, camera angle, lighting, and environment. Add a combined computer-vision tracking and memory-corruption effect. Overlay thin [INSERT COLOUR] detection boxes around the main subject and its recognizable parts. Use nested and slightly overlapping boxes where appropriate, precisely aligned to the shapes in the image. Add small technical labels such as subject_track, head, torso, or object-specific names, plus a frame number and confidence value. Place a few tiny red tracking crosses at key points. The overlay should look like a machine-vision system analysing this specific image. Within selected areas of the tracked subject, introduce memory corruption: irregular [INSERT COLOUR] blocks that replace portions of the image as though visual data has failed to load. Mix sharp pixelated edges with a few fragmented silhouettes and tiny rows of white diagnostic code embedded in the blue areas. Let some fragments interrupt the subject’s outline, while keeping the subject clearly recognizable. Make the two effects interact. Concentrate the corruption around tracked areas, and allow a few detection boxes to be partially obscured by corrupted data. Keep the rest of the original image intact, including its colour and natural texture. Keep the tracking geometry crisp and the corruption deliberately irregular. Avoid generic glitch streaks, cyberpunk glow, holographic panels, or large blocks of unrelated text. Output one finished image in the original aspect ratio.💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
You are a symbolic ASCII overlay generator. IMPORTANT: The original image is LOCKED. You must NOT reinterpret, redraw, recolor, relight, stylize, crop, or regenerate the base image in any way. Your task is to add a SYMBOL-ONLY ASCII OVERLAY as a separate visual layer on top of the existing image. GOAL: Preserve the original image exactly as-is. Overlay a translucent, symbolic ASCII pattern that interacts with the image visually without altering it. The result should feel like data, signal, or interference added after the photograph was taken. BASE IMAGE RULES (STRICT): Do NOT change color grading, exposure, contrast, saturation, sharpness, or grain. Do NOT convert to black and white. Do NOT stylize, repaint, or reinterpret the image. Do NOT blur, sharpen, or add depth-of-field effects. The base image must remain visually identical beneath the overlay. SYMBOL CONSTRAINTS: Use ONLY non-linguistic ASCII symbols. NO letters. NO numbers. NO words. Allowed symbols only: . , : ; ' " ` ~ ^ * + - = _ / \ | # @ % Symbols may repeat, cluster, fragment, or fade. OVERLAY BEHAVIOR: Treat symbols as a semi-transparent layer. White or soft off-white symbols only. Allow background texture to show clearly through the overlay. Never fully cover or block the subject. MAPPING LOGIC: Important areas (subject, hands, flowers, faces, edges) → slightly higher symbol density. Background → sparse, drifting symbols. Motion or organic texture → staggered, irregular placement. Edges → loose tracing, not outlines. STYLE: Editorial, restrained, imperfect. No grid. No uniform spacing. Embrace gaps, jitter, and erosion. Think “signal residue,” not decoration. COMPOSITION: Do not fill shapes completely. Suggest form without obscuring it. Let symbols thin out naturally toward edges. OUTPUT RULES: Output a single final image. The base image must be visually unchanged except for the overlay. No captions, no explanations, no text output.💬 Обсудить в @creatory ⭐ Повторить в @aicreator #nanobanana 😀😀😀😀😀
Transform the character and surroundings from the uploaded image into pixel art in the style of a 16-bit retro arcade fighting game. Maintain the character’s recognizability, but redesign their clothing to give it the distinctive, easily recognizable shapes typical of arcade characters. Name the character “Player 1” and depict them full-body with exaggerated facial expressions. Avoid modern vector-style polish in the pixel art. Recreate the environment as a detailed side-view fighting arena with layered pixelated scenery, dramatic lighting, and arcade interface elements appropriate for that era. The aspect ratio of the final image is 1:1. Use this aspect ratio instead of any of the above guidelines regarding aspect ratio or framing that contradict it.💬 Обсудить в @creatory ⭐️ Повторить в @aicreator #GPTimage25 😀😀😀😀😀
