| پرامپت (دستور متنی) | string | تنظیم پارامتر Text instructions describing the desired transformation. Maximum 20,000 characters. جهت هدایت دقیق فرآیند تولید مدل. | Preserve the original subject identity, facial structure, hairstyle, pose, body proportions, camera framing, and overall composition from the source image. Transform the environment into a flooded underwater railway terminal with giant glass ceilings showing whales and ocean water above. Add realistic cinematic water reflections, soft underwater caustic lighting, floating atmospheric particles, subtle wetness on clothing and skin, and highly detailed environmental storytelling elements such as abandoned luggage, vines, cracked marble, and cinematic fog depth. Maintain natural realism and believable anatomy while enhancing texture fidelity, lighting realism, color harmony, and cinematic atmosphere. Keep the face sharp and recognizable with authentic skin detail and emotionally grounded expression. Avoid over-stylization, distorted anatomy, plastic skin, blurry textures, extra limbs, cartoon aesthetics, oversaturated colors, low-detail backgrounds, or artificial-looking lighting. |
| آدرسهای اینترنتی تصاویر | array | Upload or provide input images to transform. حداکثر تا 16 images supported. | https://d3adwkbyhxyrtq.cloudfront.net/webassets/videomodels/gpt-image-2-image-to-image-in.jpg |
| نسبت ابعاد | (6 ) | نسبت ابعاد تصویر خروجی (مانند 16:9 یا 1:1). | auto |
| وضوح تصویر | (3 ) | نسبت ابعاد تصویر خروجی (مانند 16:9 یا 1:1). | 2K |
| کیفیت | (3 ) | تنظیم پارامتر Generation quality. 'low' and 'medium' use a faster, cheaper backend; 'high' uses the higher-fidelity backend. جهت هدایت دقیق فرآیند تولید مدل. | high |