Los 60 prompts de GPT Image 2.5 con más me gusta, de los 256 recopilados de creadores en X, las muestras de lanzamiento de OpenAI y repositorios de casos abiertos — cada uno con la imagen que generó, el texto completo del prompt y un enlace a su origen. GPT Image 2.5 ya está en el Canvas de Renoise, así que puedes copiarlos directamente.
Prompts library
La biblioteca
Prompts de GPT Image 2.5 junto a la imagen que produjo cada uno. Haz clic en cualquier tarjeta para leer el prompt completo, copiarlo y abrir la publicación de origen. La búsqueda y las etiquetas recorren toda la biblioteca, no solo las tarjetas en pantalla.
Stress-Test Sunburst With Six Dense Creative Briefs
a fashion campaign for an invented label, shot from a camera on the ground with a 24mm lens tilted fifteen degrees. Six people in electric blue, tomato red, lime green, hot pink, butter yellow and lilac, all posed differently, against plain cobalt sky. --- a domestic espresso machine drawn in section, 1960s service manual style, fine black line work with flat spot colour on cream paper. Fourteen numbered leader lines, a key down the right listing all fourteen parts, a scale bar in centimetres, and a title block reading FIG. 3 SECTIONAL VIEW MODEL C2. --- a frame grab from a 1994 infomercial, standard definition video with scanlines. A host in a teal blazer, a battered saucer on a rotating plinth with Antarctic ice frozen around one leg, a price starburst, a phone number, and three lines of legal small print. --- a 1960s Sunday newspaper comic strip on yellowed newsprint, six panels in two rows of three. A park ranger and a bear with a notched ear, the same campsite in every panel, dialogue specified word for word, one sound effect, and one panel with no dialogue at all. --- a cork notice board outside a Welsh village post office, overcast light, seven pinned items. A choral society flyer, a lost cat poster on yellow paper, a handwritten piano card, a bus timetable with four times, a sheepdog photo, tear off tabs, and one blank white card. --- a still from a hand drawn Japanese animated film, late 1990s cel style, painted watercolour background. A dawn duel on black volcanic sand, a woman with a drawn katana facing an older man with a sheathed one, cherry blossom blowing right to left.
Rotate a Wooden Chess Knight Through a Seamless Turntable
Create a photorealistic studio turntable video of a polished wooden chess knight. The knight has a minimalist carved horse silhouette: broad flat sides, a rounded elongated muzzle, a tiny dark eye, an angular ear, and a gently curved neck. A smooth, dark brown insert follows the mane along the back. The figure stands on a wide circular wooden base with several concentric stepped rings. Use warm brown wood with clearly visible vertical grain, softly rounded edges, and a glossy lacquer finish. Preserve the exact shape, proportions, wood grain, and dark mane insert throughout the video. The entire chess piece, including its base, rotates smoothly through one complete 360-degree turn around its vertical axis at a constant speed. It stays perfectly centered and firmly on the surface. The first and last frames match for a seamless loop. Keep the camera completely stationary, looking slightly downward at the piece. Show the entire object with a small margin above and below. Use a seamless light-gray studio background and floor, soft diffused lighting, gentle highlights on the lacquer, and a subtle contact shadow beneath the base. Duration: 3 seconds. Frame rate: 30 fps. Square 1:1 composition. No camera movement, zoom, cuts, wobbling, floating, deformation, changing proportions, sliding wood textures, flickering, additional objects, text, or logos.
A photorealistic analog-film style portrait of a young East Asian woman with pale skin and a slim figure. She has messy, slightly tousled shoulder-length black hair with long bangs falling over her right eye. She looks directly at the camera.
A photorealistic candid outdoor portrait of a young woman standing in a peaceful urban park on a sunny afternoon. She has long dark-brown hair styled in a loose low ponytail, with wispy face-framing strands gently falling across her face.
Cybernetic horror portrait, gaunt humanoid figure with cracked porcelain-white skull-like mask, mismatched hollow eye sockets (one sunken void, one recessed metallic ring), jagged exposed teeth, surrounded by a chaotic tangle of thick black cables and industrial bobbin/coil attachments wired into the head, tattered dark fabric top, dramatic low-key lighting, deep black background, high contrast monochrome, horror photography, cinematic, hyperdetailed texture, 85mm lens, shallow depth.
cinematic, high-fidelity portrait of a young woman standing perfectly still on a subway platform as a train rushes behind her. She has warm brown hair with soft bangs and a few loose strands blowing in the wind created by the passing train.
Photorealistic Editorial Portrait with Heart-Oval Face
Photorealistic editorial portrait. Knee-up cowboy shot, eye-level view. A 20-year-old adult East Asian woman, long heart-oval face with high cheeks, warm hazel almond eyes with lifted corners, slim medium-bridge nose, full coral-rose lips with a plush lower lip,
Create an ultra-realistic lifestyle portrait of a stylish adult man relaxing in a cozy upscale library lounge during the early evening. Position him seated comfortably in a vintage cognac-brown leather armchair, holding a book, with soft ambient lighting from a nearby floor lamp and a warm fireplace glow. The atmosphere is serene and sophisticated, with rich wood shelves filled with books in the background.
Create a premium contemporary editorial art poster for [LOCATION]. FORMAT: Vertical 3:4 | High resolution | Fine-art print quality | Modern art-book aesthetic Create a visually striking, instantly recognizable interpretation of [LOCATION].
Create an ultra-realistic close-up studio portrait of an adult man with a confident, thoughtful expression, photographed from the chest upward in a premium modern editorial style. Give him neatly short black hair.
Create a photorealistic editorial portrait. Full-body portrait, eye-level view. The subject is one 20-year-old Japanese or Korean female portrait subject with medium smooth hoop earrings, clean polished metal circular silhouette.
Create a vertical 3:4 luxury editorial advertisement inside a sophisticated dark emerald-green studio. Position an enormous luxury wristwatch on the right side of the frame.
Create a vertical 9:16 handheld smartphone selfie of a clearly adult young woman, 18+, standing on a sunlit urban rooftop terrace at golden hour. Crop from the top of her head to mid-thigh so the short dress is clearly visible.
Scene ID S1, location: a compact modern apartment kitchen breakfast counter. The image must clearly show a continuous kitchen countertop and a stainless steel sink, with a visible stovetop and frying pan deeper in the scene.
A polished vertical beauty portrait of a petite Spanish woman seated beside a mint-green indoor pool. Preserve the exact delicate, soft, doll-like facial proportions described below. The character occupies approximately 80
A young East Asian / Indonesian woman with a fair, smooth complexion and soft makeup (subtle blush, defined lashes, natural lip tint). She has a gentle, slightly shy smile and is looking toward the camera with a warm, friendly expression.
9:16 photorealistic candid subway portrait of an adult East Asian woman sitting by the train window at night, long dark slightly messy hair loosely tied back with soft face framing strands, natural soft makeup, clear skin texture, muted pink lips.
Create a high-fashion editorial close-up portrait of an adult man against a seamless pure white studio background. Use an intentionally asymmetric composition:
Portrait of a young East Asian woman with long, slightly wavy dark brown hair and soft bangs falling over her forehead. She wears thin gold round wire-frame glasses. She looks directly at the camera with large dark eyes, a calm,
Create a trending anime art style image from the uploaded subject. Use confident line-work with slight variation and minimal cel shading using flat shadow shapes. Use bright, saturated colors and clean graphic lighting.
Create an ultra-realistic premium lifestyle portrait of an adult man standing confidently inside a luxurious modern interior near a dark wooden doorway and a large floor-to-ceiling mirror.
A beautiful young Indonesian woman with fair skin and soft natural makeup, looking at the camera with a calm, slightly seductive expression. She has shoulder-length dark brown hair with wispy bangs, slightly wavy and naturally tousled.
A menu becomes collectible when every flavor enters through the same frame. identical black poster fields, cup scale, title zone, bottom microcopy, and brass hanging clips across the full grid. The individual drinks feel like a designed edition, not four unrelated coffee.
Without using a skill, create a sprite sheet containing the full motion of a male pixel-art goblin drawing and firing a bow, for a total of 16 frames. Then slice the sheet into frames and make a seamless looping GIF.
Cinematic close-up portrait of a young East Asian woman with shoulder-length wavy light brown hair with soft honey highlights and wispy bangs falling across her forehead, looking slightly upward and to the side with a calm,
Photorealistic close-up portrait of a young East Asian woman in her early 20s, long wavy dark brown hair with loose strands falling across her face and shoulders, wearing round tortoise-shell glasses. She is winking one eye with a soft,
9:16 photorealistic cinematic portrait of an adult East Asian woman inside a dark ancient Chinese ritual shrine, surrounded by old Nuo masks, candles, carved wooden altar objects and deep red decorations.
OpenAI lanzó ChatGPT Images 2.5 el 8 de septiembre de 2026. Dentro de ChatGPT no hay selector de modelo: generas una imagen y ya lo estás usando. En la API se publica como dos modelos, GPT-Image-2.5 Flare, el rápido, y GPT-Image-2.5 Sunburst, el de mayor calidad. No existe un único modelo llamado «gpt-image-2.5».
2.¿Qué cambió frente a GPT Image 2?
+
El control, no el tamaño. Las ediciones tocan solo lo que pediste, las ediciones en varias rondas no se desvían, los sujetos de las referencias siguen siendo reconocibles y la generación es hasta un 50% más rápida. Los límites duros del modelo de OpenAI no se movieron: hasta 4K, hasta 16 imágenes de referencia y el mismo rango de 1:3 a 3:1. OpenAI no afirma ninguna mejora en el renderizado de texto.
3.¿De dónde salen estos prompts?
+
De publicaciones abiertas en X donde alguien compartió el prompt exacto junto al resultado, de las muestras de lanzamiento de OpenAI y de repositorios de casos abiertos. Cada tarjeta enlaza a su fuente. Quedan fuera los prompts que necesitan una imagen de referencia, editar una foto existente o una conversación de varias rondas: aquí todo se reproduce solo con texto.
4.¿Puedo usar estos prompts en Renoise?
+
Sí. GPT Image 2.5 está en el selector de modelos del Canvas de Renoise, junto a GPT Image 2, Nano Banana Pro, Seedream y el resto de la línea, con una sola suscripción para todos. Copia el prompt de cualquier tarjeta, pégalo en el editor y elige GPT Image 2.5. La página del modelo tiene sus especificaciones y el precio por generación.
5.¿Cómo consigo el mismo resultado?
+
Copia el prompt tal cual y ajusta la proporción a la que muestra la tarjeta. Entrecomilla carácter por carácter cualquier texto que quieras que aparezca literal. Los resultados varían entre ejecuciones y entre modelos: la imagen de la tarjeta es una tirada, no una garantía.
6.¿Con qué frecuencia se actualiza?
+
A diario. La instantánea actual se tomó el 2026-09-11 y muestra los 60 prompts con más me gusta de los 256 recopilados hasta ahora.
7.¿Puedo aportar prompts?
+
Sí. La biblioteca se construye con publicaciones abiertas, así que compartir tu prompt junto a la imagen en X es la vía directa; el repositorio awesome-gpt-image-2-5-prompts de GitHub, enlazado sobre la cuadrícula, también acepta propuestas.
Empieza a crear con GPT Image 2.5 en Renoise
Renoise integra GPT Image 2.5 junto a Nano Banana Pro, Seedream 5.0 y el resto de modelos de imagen en el mismo Canvas: una suscripción, sin configuración, y con vídeo y audio en el mismo proyecto. Pega cualquier prompt de esta biblioteca y ejecútalo.
