Goal: Create a dark-mode AI chat screenshot demonstrating an image-generation result from a sketch prompt. The main generated artwork should show an anime schoolgirl lying on her back in a bright grassy meadow, viewed from directly overhead, smiling up at the viewer.
Canvas: Square 1:1 image, approximately 960 x 960 px, with a nearly black background like a chat interface. Use rounded corners and soft UI spacing.
Layout: In the upper-right area, place a compact user message block containing exactly 2 image thumbnails side by side: 1) a dark thumbnail with a simple white stick-figure sketch lying diagonally, and 2) a small full-body anime character reference on white. Below the thumbnails, place a rounded dark gray speech bubble containing the Japanese prompt text {argument name="chat prompt text" default="このキャラクターが草原に横たわってこちらを笑顔で見上げている。"}. On the left side beneath that, add small gray status text reading {argument name="thinking status text" default="2m 7s考えました ＞"}. Below, place the large generated image card aligned left, with rounded corners, occupying about 56% of the canvas width and height. Under the image card, add small muted chat action icons: copy, thumbs up, thumbs down, share, and ellipsis, exactly 5 icons.
Main generated image: A colorful anime illustration of {argument name="character name" default="a cute teenage girl"} lying diagonally across a lush meadow, head near the lower-right area and feet toward the upper-left. She is seen from a top-down bird’s-eye perspective, arms spread casually, palms open, legs straight, smiling brightly with an open mouth and looking directly upward at the viewer. She has short dark charcoal hair with a side ponytail or high side tuft, mint-green ribbon accents, large bright blue eyes, and round teal glasses. She wears oversized black-and-silver headphones around her neck, a light gray oversized jacket, a white blouse with a mint-green bow, a navy pleated skirt, white scrunched socks, and green sneakers. The character should be full-body but naturally integrated into the meadow, not oversized relative to the scene.
Background inside the generated image: Dense sunlit green grass with clover-like leaves, small white daisies scattered throughout, soft painterly texture, and a few blurred leafy shadows around the edges. Bright cheerful spring lighting, saturated greens and yellows.
Visual style: Clean modern AI chat UI screenshot combined with high-quality anime illustration. The chat interface is minimal, dark, and spacious; the generated image is vivid, polished, and softly shaded. Use rounded image corners and no visible watermark.
Constraints: Include exactly 2 thumbnails in the user message, exactly 1 Japanese speech bubble, exactly 1 large generated image card, and exactly 5 small action icons. Preserve the Japanese text exactly as shown in the parameter defaults unless changed. Do not add extra panels, logos, timestamps, or unrelated text.
目标：创建一张深色模式的 AI 聊天截图，演示根据草图提示词生成的图像结果。生成的主图应展示一位动漫女学生仰卧在明亮的草地上，从正上方俯视视角，微笑着看向观众。
画布：方形 1:1 图像，尺寸约为 960 x 960 像素，背景为接近黑色的聊天界面。使用圆角设计和柔和的 UI 间距。
布局：在右上角区域，放置一个紧凑的用户消息块，其中包含并排的 2 个图像缩略图：1) 一个深色缩略图，显示简单的白色火柴人对角线躺卧草图；2) 一个白色背景上的全身动漫角色参考图。在缩略图下方，放置一个圆角深灰色气泡，内含日语提示词文本 {argument name="chat prompt text" default="このキャラクターが草原に横たわってこちらを笑顔で見上げている。"}。在其左下方，添加一行灰色的小字状态文本 {argument name="thinking status text" default="2m 7s考えました ＞"}。下方放置左对齐的大型生成图像卡片，采用圆角设计，占据画布宽度和高度的约 56%。在图像卡片下方，添加 5 个小巧且柔和的聊天操作图标：复制、点赞、点踩、分享和省略号，共计 5 个图标。
主要生成图像：一张色彩鲜艳的动漫插画，主角为 {argument name="character name" default="a cute teenage girl"}，对角线躺在茂密的草地上，头部靠近右下角，脚部朝向左上角。采用鸟瞰俯视视角，双臂随意张开，手掌摊开，双腿伸直，笑容灿烂，张嘴直视观众。她留着深炭色短发，扎着侧马尾或高侧发束，带有薄荷绿丝带装饰，拥有明亮的大蓝眼睛和圆形的青色眼镜。脖子上挂着超大号黑银色耳机，身穿浅灰色大码外套、白色衬衫配薄荷绿蝴蝶结、海军蓝百褶裙、白色堆堆袜和绿色运动鞋。角色应为全身像，并自然地融入草地场景中，比例适中。
生成图像背景：阳光充足的茂密绿草地，夹杂着三叶草状叶片，散布着小白雏菊，具有柔和的绘画质感，边缘处有少量模糊的叶影。明亮欢快的春日光照，饱和的绿色和黄色调。
视觉风格：简洁现代的 AI 聊天 UI 截图与高质量动漫插画相结合。聊天界面极简、深色且宽敞；生成图像生动、精致且阴影柔和。使用圆角图像，且无可见水印。
约束：用户消息中必须包含 2 个缩略图，1 个日语气泡，1 张大型生成图像卡片，以及 5 个小操作图标。除非另有更改，否则请保留参数默认值中的日语文本。不要添加额外的面板、Logo、时间戳或无关文本。
Comments & Questions (0)
No comments yet
Be the first to comment!