GPT-image-2 API d’images IA
GPT Image 2 est le modèle d'image d'OpenAI exposé dans le flux de génération d'images Flatkey. Utilisez /v1/images/generations et configurez le nombre, la taille, la qualité, le format, l'arrière-plan et la modération avant l'envoi.
Sortie
Aperçu
Résumé de la requête
- Point de terminaison
- POST /v1/images/generations
- ID du modèle
- GPT-image-2
- Format
- 5:7
- Résolution
- 1024x1024
- Qualité
- high
- Sorties
- 1
Aperçu uniquement. Connectez-vous pour envoyer cette requête à Flatkey.
{
"model": "gpt-image-2",
"prompt": "Design a high-end e-commerce pastry packaging solution for [Brand Name] that can be mass-produced. The product name is [Product Name], the product type is [Pastry Type], the brand positioning is [Brand Positioning], and the target audience is [Target Users]. Packaging types include [Heaven-and-Earth Lid Gift Box / Drawer Box / Window Color Box / Handheld Gift Box / Magnetic Gift Box / Color Box / Paper Tube / Soft Bag / E-commerce Shipping Gift Box]. The finished product dimensions are [Length × Width × Height mm]. Internal specifications are [quantity] pieces or [combination pack], and each piece is specified as [Single Weight or Size]. \n\nPlease design a professional-grade high-end e-commerce pastry packaging design proposal board. Overall, it must meet international packaging design award-level standards, combining brand recognition, main e-commerce image appeal, shelf display effect, gift value, structural rationality, printability, transportation protection, and factory mass production feasibility.\n\nThe overall style is [Oriental New Luxury / Minimalist and Sophisticated / Natural Organic / Young National Trend / New Chinese Elegance / Vintage Collection / Holiday Gifts / Light Luxury Souvenirs / French Dessert Style / Japanese Minimalist / Regional Culture Style]. The visual temperament should be high-end, refined, clean, appetizing, and have brand memorability, avoiding a sense of cheapness, clutter, and template-like feeling. The color scheme should focus on the main color, paired with auxiliary colors and accent colors. The overall color should match the product's taste, brand positioning, and gifting scenario.\n\nThe image is a horizontal professional packaging design proposal board, clearly divided into zones, with advanced layout, reasonable white space, and complete information. It includes: main visual 3D packaging rendering, front of packaging, back of packaging, left left, right right, top top, bottom bottom, internal structure of unboxing, cake tray display, complete unfolding knife pattern diagram, dimensioning, line pattern legend, material description, printing process description, partial process enlargement drawing, color scheme, font scheme, and manufacturer production information.\n\nThe main visual 3D packaging rendering should show the packaging box from a 45° stereoscopic perspective, with authentic packaging materials, delicate paper texture, clear box thickness, reasonable structure, and soft, high-end light and shadow. Surrounding items can be arranged with real pastries, ingredients, tea sets, plates, wooden trays, ribbons, flowers, grains, nuts, fruits, cream, sesame, egg yolks, red beans, cocoa, honey, wheat ears, festive elements, or regional cultural elements to create a high-end e-commerce main image. The color and shape of the pastries must match the specific product type and are not limited to green.\n\nThe front design must include the brand name, product name, English auxiliary name, core selling points, net content, specification information, brand symbol, and main visual design. The main visual can use pastry photography, illustrations, patterns, window structures, oriental graphics, regional cultural symbols, raw material graphics, abstract brand symbols, or holiday designs. The composition needs to be strongly memorable, suitable for e-commerce thumbnail clicks.\n\nThe back design must include the product name, ingredient list, allergen warnings, nutrition facts sheet, production information, storage methods, consumption recommendations, standards, production license number, manufacturer information, barcode, QR code, environmental label, and a reserved area for the production batch number. The information hierarchy is clear, and the layout is authentic and trustworthy. The left and right sides should include the brand slogan, product series name, flavor label, unboxing direction, decorative patterns, or brand auxiliary graphics. The top displays brand symbols, gift box opening methods, sealing structures, or main visual extension patterns. The bottom displays the folded structure, production batch number location, environmental label, or simplified information area.\n\nThe internal structure display must include food-grade PET trays, eco-friendly pulp trays, paper card partitions, individual small boxes, individual packaging bags, or greaseproof paper linings. Each pastry is placed independently, ensuring shock resistance, pressure resistance, flavor transfer, oil stain resistance, and e-commerce transportation safety. The pastries have realistic shapes and appetizing appeals, with naturally sophisticated colors.\n\nThe unfolding tool diagram must show the complete packaging unfolding structure, including front, back, left and right sides, top, bottom, tongue insert, latch, adhesive position, window area, internal tray diagram, barcode QR code position, size marking, and process layout. The die diagram must resemble the actual packaging engineering drawing, with accurate proportions and clear lines, making it suitable for manufacturer sample reference.\n\nThe die drawing must be clear, professional, and mass-producible, with all units in mm. Line type legends must be annotated: Red solid line = cutting line / cutting line; Blue dashed line = indentation line / polyline; Green line = bleeding line; Purple line = safety line; Gray line = reference line, not printed; Yellow area = adhesive position; Translucent light blue area = food-grade PET window area; Gold area = hot stamping edition; Bright white area = partial UV version; Embossed shadow area = bump / recess area.\n\nSize standard: bleeding 3 mm per side; Safety margin 5 mm; Dimensional tolerance ±1 mm; Die-cutting tolerance ±0.5 mm; Printing nest tolerance ±0.3 mm; Barcode areas must be clearly marked with a white background; QR codes must be reserved in the scanning safe area; The food label information area must be clear and legible; All folds, tongue inserts, adhesive points, and window areas must be clearly marked.\n\nMaterial recommendations: [350g food-grade white card paper / 400g food-grade white card paper / specialty paper / grayboard mounting / kraft paper / eco-friendly pulp / biodegradable paper tray / food-grade PET / food-grade oilproof paper / aluminum foil inner bag / individual small packaging]. The food-contact section must use food-grade materials, and e-commerce transportation must consider pressure resistance, shock resistance, moisture resistance, oil stain resistance, and odor transfer.\n\nPrinting method: CMYK four-color printing, with Pantone spot colors, metal spot colors, and special plate processes available. Surface Processing: [Matte Film / Glossy Film / Tactile Film / Localized UV / Hot Stamping / Silver Hot Stamping / Rose Gold Hot Stamping / Embossing / Embossing / Embossing / Pearlescent Ink / Spot Color Printing / Food-Grade Eco-friendly Ink].\n\nThe partial process enlargement should display the logo stamping, brand name accents, product name partial UV exposure, paper embossing, illustration details, window PET frame, lock structure, inner tray details, and food label information area. The color scheme should display the main color, secondary color, accent color, and indicate the Pantone color code, CMYK value, RGB value, or approximate color value. The font scheme should display the brand's Chinese font, product name font, English auxiliary font, numeric font, and information font.\n\nOverall image requirements: award-winning pastry packaging design, premium bakery packaging, high-end confectionery packaging, luxury e-commerce dessert gift box, production-ready packaging blueprint, manufacturer-ready dieline layout, structural packaging engineering drawing, realistic paper texture, refined typography, premium brand identity system, elegant food packaging design, photorealistic 3D packaging mockup, detailed printing process annotations, CMYK print-ready layout, Pantone color system, gold foil stamping, embossed logo, debossed texture, spot UV details, food-grade PET window, inner tray structure, sustainable food packaging, premium gift box design, clean white studio background, soft natural lighting, ultra detailed, high resolution, commercial product design board",
"n": 1,
"size": "1024x1024",
"quality": "high",
"output_format": "png",
"background": "opaque",
"moderation": "auto"
}capacités
GPT Image 2 performances et disponibilité de l'API
La télémétrie des requêtes Flatkey pour GPT Image 2 apparaît lorsque le trafic est suffisant ; aucun benchmark ni score de qualité n’est déduit.
API
GPT Image 2 utilisation de l'API et activité des requêtes
Cette section reflète les requêtes réelles Flatkey pour GPT Image 2 et reste sans données tant qu’un trafic suffisant n’est pas réuni.
Tarifs
Tarifs de GPT Image 2 par token d'entrée image
Les prix ci-dessous sont calculés à partir des données tarifaires Flatkey pour ce modèle et des groupes visibles renvoyés par notre API de tarifs.
Ajouter des crédits
Utilisez le même solde et la même clé API Flatkey pour les modèles image, vidéo, audio et texte.
- Type de modèle
- Texte vers image
- API
- /v1/images/generations
- Base de facturation
- image
- Sorties
- 1–10
Capacités de génération et d’édition d’images de GPT Image 2
Les champs d'intégration documentés sont présentés ci-dessous ; aucune promesse de benchmark ou de qualité n'est déduite.
Texte vers image et édition
Créez de nouvelles images ou retouchez une image via /v1/images/generations et /v1/images/edits.
Création guidée par références
Utilisez le contexte visuel pour préserver sujets et composition dans les travaux produit et campagne.
Variantes visuelles flexibles
Déclinez un brief en formats carré, portrait ou paysage pour chaque emplacement.
Livraison sûre pour la production
Préparez des résultats selon les comportements documentés de qualité, format, fond et modération.
Comparer
GPT Image 2 face à GPT Image 1 : champs de migration
Les champs d'intégration documentés sont présentés ci-dessous ; aucune promesse de benchmark ou de qualité n'est déduite.
| Capacité | GPT Image 1 | GPT Image 2 |
|---|---|---|
| Point de terminaison API | Inconnu ici | /v1/images/generations; /v1/images/edits |
| Tailles | Inconnu ici | 1024x1024, 1536x1024, 1024x1536, auto |
| Formats | Inconnu ici | PNG, JPEG, WebP |
| Qualité/arrière-plan/modération | Inconnu ici | Documenté ci-dessus |
| Classement de qualité d'image | Non affirmé | Non affirmé |
Choisir la bonne requête avant l’intégration
Vérifiez les workflows, le coût actuel et le contrat exact des paramètres Flatkey.
Calculée avec les réglages actuels et le tarif du catalogue lorsqu’un prix unitaire fixe existe.
- Configuration actuelle
- 1024x1024 · high · 1×
- Tarif en direct
- $0.0088 / image
- Requête estimée
- $0.0088
Le montant final dépend de la requête et du groupe de compte ; les tarifs par jetons ou paliers ne donnent pas un montant fixe.
Workflows et Model ID
Utilisez uniquement les workflows disponibles. Flatkey conserve un ID public et le corps choisit l’entrée.
Texte vers image
Générez une image depuis un texte.
- ID du modèle
gpt-image-2- Endpoint
/v1/images/generations
Image guidée par référence
Utilisez des références pour modifier ou varier une image.
- ID du modèle
gpt-image-2- Endpoint
/v1/images/generations
Compatibilité des paramètres
Valeurs issues de la même configuration que le Playground.
nImages1–10
Par défaut: 1sizeTaille1024x1024 · 1536x1024 · 1024x1536 · auto
Par défaut: 1024x1024qualityQualitéauto · high · medium · low
Par défaut: highoutput_formatFormat de sortiepng · jpeg · webp
Par défaut: pngbackgroundArrière-planopaque · auto
Par défaut: opaquemoderationModérationauto · low
Par défaut: autoimageLimite de références4 fichiers
Endpoint: /v1/images/generationsDu premier test à la production
Gardez le même contrat dans le Playground et l’API et gérez validation, solde, état et résultat.
Cycle de la requête
- 1
Valider modèle, prompt, références et sortie.
- 2
Envoyer la requête image.
- 3
Lire le résultat ou la tâche asynchrone.
- 4
Stocker le fichier et ses métadonnées.
Contrôles de production
Bloquez les valeurs ou références non prises en charge.
Gérez séparément clés, solde, limites et groupe.
Identifiez le prompt ou le média à remplacer.
Conservez le task ID, limitez les reprises et ne marquez pas un délai comme succès.
Bibliothèque de prompts
Exemples de prompts GPT Image 2 pour produits et marketing
Commencez par un sujet, une action, une caméra ou une composition précise et un réglage de sortie explicite ; séparez les champs de requête du prompt.
Généré avec Image 2
Scene: A 16:9 sci-fi game loadout screen in a blue-purple spaceship hangar. A female armored character stands centered with a glowing violet rifle; armor slots line the left, three weapon slots line the right, and the lower slot is highlighted. Keep the HUD hierarchy, pose, lighting, and equipment geometry fixed for an exact equipment-switch frame. No readable words, logos, invented stats, or watermark. Composition and rendering: Deliver one 16:9 game-interface keyframe. Separate the character, held equipment and HUD into clear depth layers; keep slot spacing and icon scale consistent, with the selected item visually dominant. Preserve the scene-specific game art style. Let material edges catch the existing light without letting interface glow obscure the silhouette. Keep all equipment fully inside the frame. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Généré avec Image 2
Scene: A rainy night soccer broadcast frame under bright stadium floodlights. A black-uniform player strikes the ball while two white-uniform defenders close in, with the goal and a blurred crowd behind. Keep the broadcast camera angle, blank blue score bars, spray, and motion blur consistent; leave overlays abstract and text-free. No real teams, athletes, logos, or watermark. Composition and rendering: Deliver one 16:9 broadcast still at the decisive instant. Use coherent perspective for the playing surface and equipment, readable separation between competitors, and a softer crowd behind the action. Keep the principal subject sharp, with directional blur limited to fast extremities and spray. Preserve the specified broadcast angle and reserve overlay areas clear of the action. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Généré avec Image 2
Scene: A closed matte-black square smartwatch with a black strap rests on a wet glossy black tabletop. Water droplets catch a cool blue rim light and a soft reflection sits beneath the watch. Use a low three-quarter product camera and preserve the case, strap, highlights, and empty dark background. No readable branding, extra products, hands, or watermark. Composition and rendering: Deliver one 16:9 commercial product still. Keep the complete product silhouette inside generous crop-safe margins. Resolve edges, seams and material transitions precisely; shape highlights to describe volume without clipping bright surfaces or crushing dark details. Align reflections with the object and light source. Retain the stated camera angle and existing negative space; add no decorative props. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Généré avec Image 2
Scene: On a rain-soaked observatory roof at sunrise, a lone figure in a long dark coat walks toward a large telescope beside an open dome. Wet stone reflects the warm doorway light; cloud-covered mountains sit beyond. Hold the screen direction and end on the telescope and figure in the same wide composition. No readable text, logos, or watermark. Composition and rendering: Deliver one 16:9 storyboard keyframe, not a collage or an action sequence. Separate foreground, character plane and background through the existing light and atmospheric depth. Keep the stated pose and eyelines readable, and retain environmental landmarks for continuity. Match the described photographic, illustrated or puppet medium; give fabric and surfaces texture appropriate to that medium. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Généré avec Image 2
Scene: In a bright home kitchen, a surprised cook in a blue shirt and cream apron reaches toward pancakes, a frying pan, flour, bowl, and whisk suspended midair. Freeze the comic cause-and-effect moment with believable weight, warm daylight, and a clear path for each object. No injury, logos, readable words, or watermark. Composition and rendering: Deliver one 16:9 physical-comedy still at the described instant. Make the cause of the gag readable through body balance, grip and object spacing. Separate the reaction from the airborne props; retain believable gravity and contact points. Keep the face and important props sharp, with only restrained directional blur on fast edges. Preserve the stated lighting and room or stage layout. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Généré avec Image 2
Scene: A rainy period harbor platform shelters people in era-appropriate coats and umbrellas as a vintage green tram arrives on wet tracks. Wooden waterfront buildings and misty mountains sit behind the reflections. Preserve the historical clothing, tram shape, rain direction, and stable wide composition; no readable signage, logos, or watermark. Composition and restoration: Deliver one 16:9 archival-style still. Preserve the original framing, period-specific construction and restrained tonal range; retain monochrome, sepia or faded color as described. Recover modest detail in faces, clothing and architecture without synthetic sharpening. Keep fine grain and authentic aging consistent across the frame; repair damage without replacing historical features. Avoid: warped perspective, malformed anatomy, merged objects, inconsistent shadows, excessive sharpening and watermarks. Follow the scene-specific text and branding restrictions above.
Pourquoi Flatkey
Pourquoi utiliser GPT Image 2 avec Flatkey ?
Gardez visibles les faits, les champs de requête et les limites tarifaires de GPT Image 2 du test à la production.
Tous les contrôles documentés
Définissez n, size, quality, output format, background et moderation avant de transmettre la requête à la console.
Tarification explicite
Examinez chaque dimension documentée d'entrée, de sortie et de cache de GPT Image 2, plutôt qu'un prix unique présenté comme universel.
Départs adaptés au workflow
Utilisez les exemples produit, publicité et storyboard comme prompts modifiables.
Aucune promesse non vérifiée
La page ne promet ni sortie transparente, ni génération gratuite, ni prix fixe par image lorsque ces faits ne sont pas vérifiés.
API
Endpoint API et configuration de requête de GPT Image 2
Envoyez l'ID du modèle avec les contrôles d'image pris en charge.
curl https://router.flatkey.ai/v1/images/generations \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-***" \ -d '{"model":"gpt-image-2","prompt":"Creative direction: Create a vertical illustrated fantasy narrative poster with a large dark-haired profile silhouette on the left, a layered mountain temple city in the center, a red torii gate, drifting cherry blossoms, and small travelers along the bottom edge. Use a cream paper texture with ink-and-wash detail, a restrained charcoal, vermilion, and muted violet palette, and reserve a clean right margin for the title and credits.\n\nComposition, material and finish: Build the silhouette and temple city as nested visual layers, with the gate as the vermilion focal accent. Keep small travelers distinct from architectural detail. Use delicate ink edges and paper grain rather than photographic skin or glossy 3D surfaces. Leave the right title margin genuinely empty; avoid added lettering, duplicated gates, muddied silhouettes and watermarks.","n":1,"size":"1024x1024","quality":"high","output_format":"png","background":"opaque","moderation":"auto"}'
Questions fréquentes
GPT Image 2
API : tarifs et questions de prompts
Tarifs, compatibilité, limites et détails de requête propres au modèle.





