Jack

hello！

[Jack @EhcoRelay]

@thinkszyg Good job.

[Jack @EhcoRelay]

I love how ChatGPT Work turns ideas into finished work. I used it to build “拍同款”, a browser extension that turns any web image into an editable AI prompt and helps create your own version—all in one flow. Less switching, more creating. #ChatGPTWork https://t.co/rr7IxNRna9

[Jack @EhcoRelay]

Thanks for sharing! This style looks super cool. I actually created a similar page myself and published it using ChatGPT's site feature — here's the link: https://t.co/VE9SkrlKDW

[Marina Budarina @marina_uiux]

some new design experiments of mine WIP

[Jack @EhcoRelay]

Cool

[Virena @vireonixx]

GPT Image 2 on ChatGPT.

Prompt: Create an ultra-detailed retro Japanese anime portrait inspired by authentic late-1980s and early-1990s shoujo cel animation. Illustrate a graceful young woman with luminous porcelain skin, large icy blue-gray eyes, long feathered lashes, subtle rosy blush, glossy scarlet-red lips, and a voluminous jet-black chin-length vintage bob with sculpted waves and polished cel highlights. Dress her in a timeless black-and-white Breton striped knit top finished with a vivid red neckline, complemented by glossy red oval drop earrings. Pose her in a refined head-and-shoulders portrait with a calm direct gaze and elegant posture. Place her against a clean muted slate-blue studio background illuminated by soft diffused fashion lighting that creates smooth skin tones, glossy hair reflections, authentic vintage cel shading, crisp expressive ink linework, subtle nostalgic paper grain, rich editorial color harmony, and beautifully restored ultra-detailed 8K classic shoujo anime quality.

[Jack @EhcoRelay]

I didn't use the original creator's prompt. I reverse-engineered it from the image and regenerated it. Not too bad, right?

[draco @DracoVibeCoding]

当动漫女孩儿们称为贤妻良母~
请生成4张独立9:16图片（不要拼接）：
比例：9:16
风格：东亚都市屋顶晾衣绳纪实摄影 + 午后硬光与布料透光 + 远处电线与水塔（生活纪实摄影，非插画）
主题：动漫女角色「屋顶收衣服」
人物形象须与其在原作动漫中的形象高度一致，关键发型/发色/五官气质/标志特征清晰可辨，避免换成认不出的路人脸。
人物（每张一个，成年，东亚审美）：
1) 你的名字中的宫水三叶（成年再演绎）
2) 孤独摇滚！中的伊地知虹夏
3) ONE PIECE中的娜美
4) NARUTO中的春野樱（粉发绿眼）
服装：大胆清凉家居/夏日便装（吊带、短裤、拖鞋），生活感，非色情
构图：半身到腰，晾衣绳与衣物入镜，城市屋顶背景
约束：强纪实摄影；可识别；禁止国旗/旭日

[Jack @EhcoRelay]

I'm also into these wuxia-style clothing pieces that you could actually wear out in real life — if someone designs them, I'd love to give them a try. The prompt is in the ALT text.

[VoxCat @VoxcatAI]

这组想做的不是传统古装侠女，而是 普通女孩用低成本单品穿出江湖感。

白衬衫、旧半裙、围巾、腰带、窗帘布、编织袋边角料，重新做层次组合，再加上宽腰封、垂坠布条、旧靴子和手工发饰，画面立刻就有了新中式武侠的味道。

我喜欢这种方向，因为它重点不在“贵”，而在“会搭”。
不用复杂场景，也不用完整古装，老小区楼梯、灰墙巷口、城市角落，就能拍出一种清爽、灵动的侠气。

[Dilum Sanjaya @DilumSanjaya]

Been thinking about sharing some fun, interactive science app ideas

Made this one today

UI design and planet textures
GPT Images 2

Code
Gemini 3.1 Pro

[Jack @EhcoRelay]

@Goodmanprotocol Amazing

[Jack @EhcoRelay]

I love this vibe

[simeon-sanai @Naiknelofar788]

GPT image 2 on ChatGPT

Prompt:
Create a stylish modern travel poster inspired by risography about DUBAI. Use simplified illustrated images of the city: the Gwanganje Bridge, ocean waves, seagulls, beach umbrellas, fish baskets at the market, neon street signs and small abstract decorative figures. Create a bold, multi-layered risographic poster with imperfect ink design, translucent colors printed on top, graininess in semitones, rough paper texture, flat graphic shapes, playful geometric shapes, and energetic editorial composition. Use fluorescent pink, bright blue, warm yellow, orange-red, leafy green, and rich cobalt blue on a white paper background. Add bold font in English: Main title: “DUBAI” A subtitle describing the city.The text in the footer: ”ARTISTIC PRINTING · COASTAL STORIES · 2026“ The design should be bright, cultural, modern, playful and collectable, like the original independent risographic poster for a local creative festival. Avoid copying any existing poster, artist's style, corporate identity, logo, mascot, copyrighted character, exact layout, or recognizable commercial design. Create an original composition inspired only by risographic printing technologies and modern independent poster design.
generate for istanbul

[Jack @EhcoRelay]

@marina_uiux These look fantastic!

[Jack @EhcoRelay]

@jeff11224669 @alisaqqt It depends on which models you use. Most of the cost comes from the video generation model. The default setup uses Gemini Omni, and these two videos cost me around $10 to make. I later switched to Seedance 2.0 Mini, which brought the cost down to about $4–5.

[Jack @EhcoRelay]

Just installed the Vox Director skill in Codex. One single prompt → complete Vox-style video. The output is ridiculously good. Mind officially blown. Massive thanks to @alisaqqt for the share!

[Alisa Qian @alisaqqt]

I made a Vox-style explainer about Tang Dynasty. Just one prompt, 100% AI / end to end.

keyframes (Seedream 5.0 Pro) → animation (Omni Flash) → voice (xAI TTS) → music (MiniMax Music 2.6) → assembled locally on Claude Code.

The whole pipeline is on @atlas_cloud_ai. One topic in, finished film out.

[Jack @EhcoRelay]

Made another short video with Codex & 🎬 Vox Director— really loving this vibe.

[Alisa Qian @alisaqqt]

Hope to see more showcases!

[Jack @EhcoRelay]

I feel like I've been having trouble focusing lately and get easily distracted.