Clara998 泽佳

[Clara998 泽佳 @Zejia98]

clone

[Zed @ZedNilm1]

this brand is printing 7 figures per month

not with “original” creatives
just copying what is already proven to convert

- same hook
- same angle
- same structure

people think ad performance comes from creativity
but it actually comes from pattern recognition

you don’t need to reinvent anything

you just need to deploy what already works

i made a short breakdown on how to legally model winning ads

rt + comment “clone” and i’ll send it
(follow for dm)

[Clara998 泽佳 @Zejia98]

@NanFlorin2 122K

[松果先森 @songguoxiansen]

谷歌又又又放大招了！

谷歌发布了新的Nano Banana（小香蕉）就像就像是功能更强大的Photoshop，而且还免费。

它可以让你做到以前不可能的疯狂事情，比如：摘小姐姐的口罩、保持角色一致、多图融合、线稿上色、动漫制作、风格迁移和转换、图片推理

以下是我找到的20多个最酷的例子：

另外由于Nano Banana太热门了，它已经在Google Ai Studio正式成为模型名称。Gemini 2.5 Flash Image Preview已沦为别名。

[吕立青_JimmyLv 2𐃏26 @Jimmy_JingLv]

100 个 🍌 Nano Banana 提示词大合集，值得收藏！

我 fork 了 @AIJamezBondos 的 Awesome GPT-4o Images 仓库，并写脚本让 gemini-2.5-flash-image-preview 全部重新跑了一���。

发现确实 Gemini 是图像编辑之神，但是文生图美感不如 GPT

狠狠花钱做了对比图，求个 Star 🌟 https://t.co/JG61kCPNk3

[Linus ✦ Ekenstam @LinusEkenstam]

Googles 🍌 Nano Banana is magical

But we've just scratched the surface of whats possible

This is the post Nano Banana era, there is a clear before and after.

1. Taylor Swift wearing a random flat lay Harpers Baazar Pinterest outfit.

[Shruti @heyshrutimishra]

🚨 Google quietly dropped a VIDEO AI bombshell just 2 hours ago.

People are calling it Photoshop + TikTok mashed into Workspace.

This is going to change FUTURE of content forever: 👇

[雲風Yunfeng @YunfengCO]

#nano_banana #gemini #手办
Google gemini 2.5新出的Nano-Banana模型，手办效果杀疯了！爱豆、动漫角色...配合3D打印机会怎样？提示词在评论区📷

[Tom Dörr @tom_doerr]

real-time face swap for streaming or video calls

[Min Choi @minchoi]

China's Zhejiang University researchers just introduced TINKER.

This AI edits 3D from 1-2 photos and stays consistent from every angle, no retraining.

Any view to video.

4 wild examples + how it works:

[el.cine @EHuanglu]

this is over.. for traditional film studios

Higgsfield just dropped Product-to-Video, you can now drop any product into any environment and.. the actor can even interact with it

this is best way to do product commercial with AI.. for free

here's how:

[GitHubDaily @GitHub_Daily]

知名网页抓取工具 Firecrawl 团队，最近开源了一个功能强大的项目：Open Lovable，短短几天已狂揽了 10.9k Star。

只需输入我们想克隆的任何网站链接，它就能在几秒内基于 React 重新生成一个一模一样的网站。

GitHub：https://t.co/WdzxsM2Yrm

主要特性：

- 一键克隆任何网站为 React 应用，支持复杂布局和交互
- 支持 OpenAI、Anthropic、Gemini、Groq 等 AI 模型
- 基于 Firecrawl 抓取网页结构，准确还原样式与内容
- 支持本地调试与部署，可实时预览生成结果
- E2B 沙盒环境提供安全的代码执行和测试环境

项目基于 MIT 协议完全开源，只需克隆项目代码到本地，配置好相关的 API 密钥即可使用。

[Based Labs AI @BasedLabsAI]

AI videos are taking over — creators are racking up millions with hyperrealistic content.🎥💥

All of it possible with BasedLabs AI.

Here are 7 crazy examples:

1. Dog and lion door cam

[Higgsfield AI 🧩 @higgsfield_ai]

The next-level & fully refined Multi-Reference is now in Higgsfield.

Up to 4 reference images. Perfect character consistency. No more random artifacts.

Retweet for a FULL Multi-Reference Playbook in your DMs.

Be the first to try it:

[@levelsio @levelsio]

✨ Used Claude Code to make this new UGC talking video feature on Photo AI

Lots of people make AI UGC videos now to make ads, but they're usually random people

With Photo AI, you can make UGC videos that actually look like you or models/influencers you work with directly

Pick any voice from @ElevenLabs or use your own when you bring your own API key

I already had a talking video feature but it wasn't so good because it was stuck with only a male or female voice and lipsync tech wasn't so good yet (6 months ago), now it's becoming much better

As always in AI every few months all the models gets so much better and my app's features directly benefit from that

[Ray (movie arc) @rayisdoingfilm]

Runway Aleph experiment: using one initial vid clip as a character 'video reference' to generate many more with consistent aesthetics/world

Whole video is made from aleph generations based on one initial clip (prompt and clip in the comments)

[Dreamina AI @dreamina_ai]

Image 3.1 is here! Powered by seedream 3.0, Image 3.1 is a fine-tuned version designed sand it's our most visually expressive model yet.

Let’s break down what’s new (and what to watch out for) 👇
A full prompt guide will be attached at the end for creators who want to dive deep.

[apolinario (poli) @multimodalart]

yay! new model from @bfl_ml 🤝 @krea_ai just out

FLUX Krea [dev] focus on having its distinct aesthetics, and it shows taste also by releasing open weights & demo on @huggingface 😎

[Christopher Fryant @cfryant]

I spent the last day trying every experiment I could think of with @runwayml's Aleph video editing AI.

These are the best results.

Drop your experiments in the replies!

1/15: Gangsta style with face tattoos

[fofr @fofrAI]

Image: Seedream 3
Audio: ElevenLabs v3 alpha
Video: OmniHuman

https://t.co/ZcgxRLFQo3

[Paul from DentroAI @paul_dentro]

@IamEmily2050 @fofrAI just use vace fusionX for joint audio