GPT-Image V2 Feature Announcement

A simple instruction to generate an image illustrating the full release of the GPT Image V2 model.

适合作为创意起点：保留它的主体、构图和风格骨架，再替换成你的主题与使用场景。

画幅建议

1:1 或 4:5

细节策略

先快速出图，再细节精修

生成流程

先用文字生成定主画面，再一次只改一个重点细节。

完整 GPT Image 2 提示词

English prompt

Help me generate an image stating that {argument name="announcement content" default="GPT Image V2 is now fully open"}

Original prompt

帮我生成一张图片，说明{argument name="公告内容" default="GPT Image V2已经全面开放"}

A realistic prompt for generating an ordinary, imperfect candid shot as if taken on an iPhone, featuring natural motion blur and uneven lighting.

A creative prompt designed to have the AI personify itself and generate a realistic, candid selfie as if taken haphazardly with an iPhone.

A prompt that transforms a pet photo into a cute, hand-drawn Japanese journal-style illustration with stickers, annotations, and a healing aesthetic.

A simple and effective command for professionally redesigning an existing image using GPT Image 2.

A dynamic street photography prompt featuring a woman in a low-angle pose with iPhone aesthetics.

Transforms a reference photo into a delicate white-line illustration with faint watercolor tones while preserving the original pose and composition.