Skip to main content

🎉 Limited time: get 100 trial credits and run GPT Image 2 for just 2 credits per generation — roughly 50 flagship image runs. Try GPT Image 2

VibeArt
VibeArt

英语学习卡片

View Only

This canvas explores AI-generated English learning cards, featuring four realistic, everyday vertical 9:16 photo scenes. Each image is designed to identify five objects, annotated with English words, simple definitions, and optional example sentences using thin white lines. The overall aesthetic aims for a social media-friendly light learning style, created using text-to-image models.

English learning
AI generation
realistic scenes
annotated objects
educational
social media style
text-to-image
@cheerselflinPost 050648...Open on X
Prompt (Original)
【任务目标】在照片中识别5个物品，给每个物品加英文单词+简单释义+可选例句；白色细线手绘标注；自然留白；整体像社媒轻学习图片；比例9:16。
Double-click to edit
Prompt (Refined Text2Image)
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy.
Double-click to edit
Prompt Variant A
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant A: warm morning light, cozy home mood, simple composition with clear object separation.
Double-click to edit
Prompt Variant B
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant B: bright neutral daylight, slightly more editorial framing, stronger contrast between objects and background.
Double-click to edit
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant A: warm morning light, cozy home mood, simple composition with clear object separation.
Gemini 3.1 Flash
4/24
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant B: bright neutral daylight, slightly more editorial framing, stronger contrast between objects and background.
Gemini 3.1 Flash
4/24
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant A: warm morning light, cozy home mood, simple composition with clear object separation.
GPT Image 2
4/24
Text-to-image only (no input image). Create a realistic everyday vertical 9:16 photo scene (e.g., breakfast table by a window) containing many objects, then annotate exactly 5 common objects only. For each selected object, add white thin hand-drawn outline and a casual handwritten English note with: (1) word, (2) beginner-friendly short meaning, (3) optional short example. Keep notes natural, light, and breathable with whitespace; do not overcrowd. The style should feel like social-media "learn English from daily life" content, clean and relaxed, not dictionary-like, not decorative-heavy. Variant B: bright neutral daylight, slightly more editorial framing, stronger contrast between objects and background.
GPT Image 2
4/24
70%