Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Min Choi Profile picture
Min Choi
@minchoi
AI Educator. 𝕏 about AI, solutions and interesting things. Showing how to leverage AI in practical ways for you and your business. Opinions are my own.
25 subscribers
Sep 3 • 12 tweets • 3 min read
Nano Banana is insanely impressive.

People are unlocking new creative ways to use it.

10 wild examples: 1. Take you across Middle earth
Aug 31 • 9 tweets • 3 min read
Lovart AI + Nano Banana is pretty wild.

Drop your product + goal, and AI design team does the rest.

Nano Banana is FREE this Labor Day weekend.

6 wild examples + how to try: 1. Interior design
Aug 30 • 11 tweets • 3 min read
Less than 48 hours ago, xAI dropped Grok Code Fast 1.

And it is already breaking brains.

The speed is unreal. And it is FREE until September 10th.

10 wild examples + how to try:

1. Building battle simulator prototype
2. By far the fastest tested
Aug 29 • 11 tweets • 3 min read
Google's Nano Banana is pretty wild.

People just dropped some new insane use cases

100% AI bananas

1. What is going on here? Image 2. PROMPT: "show woman wearing the outfit"
Aug 28 • 13 tweets • 3 min read
Google Nano Banana just broke the Internet yesterday

People are getting crazy creative with it.

10 wild examples: 1. Annotate information about real world image
Aug 26 • 7 tweets • 3 min read
We are cooked.

HeyGen just dropped Digital Twin, powered by Avatar IV.

AI that makes a lifelike avatar of you.

People won't be able to tell which is real you.

Bookmark this, here’s how to set it up: Step 1. Capture.

Pick a quiet, well-lit spot. Neutral background. Place the camera at eye level.

Speak naturally for a short capture, clear diction, varied pace, a few head turns and blinks.

That diversity helps the twin learn your timing and micro expressions.
Aug 23 • 11 tweets • 3 min read
This is wild.

Mirage 2 just dropped: a generative world engine.

This AI turns any image into a playable world, then prompt to reshape it live 🤯

9 wild examples + how to try right NOW:

1. Cyberpunk city -> Rainforest -> Mountaintop castle 2. The starry night
Aug 19 • 9 tweets • 3 min read
Higgsfield AI draw-to-video is pretty wild.

Anyone can drop any image or product and bring it to life inside any video.

7 wild examples:

1. Add anything to any video 2.
Aug 15 • 12 tweets • 3 min read
Higgsfield AI absolutely cooked with Product-to-Video🤯

Now anyone can Instantly add products into cinematic shots.

10 wild examples:👇

1. It still won't save Jack 2.
Aug 13 • 11 tweets • 3 min read
This is wild.

Higgsfield AI's new Draw-to-Video lets anyone sketch on any image and watch it come to life.

RIP complex prompting.

9 wild examples + how to try:

1. Change any scenes from still image 2. Change any scenes from still frame
Aug 11 • 12 tweets • 3 min read
Massive AI news happened this past week.

Here's what you don't want to miss: Image 1. Google DeepMind dropped Genie 3

AI that can generate interactive world in real time from text, images, videos. A step closer to Holodeck
Aug 8 • 12 tweets • 3 min read
Less than 29 hours ago, OpenAI dropped GPT-5.

And some people are calling it the best model.

Some not so much.

10 wild examples: Image 1. GPT-5 refactored entire codebase in one call
Aug 6 • 12 tweets • 3 min read
It's only been just over 19 hours since OpenAI dropped gpt-oss, fully open source model finally.

And people are already getting crazy creative and results with it.

10 wild examples: 1. Building gpt-oss Pro Mode

Basically o3 Pro but open source
Aug 5 • 12 tweets • 3 min read
Higgsfield just dropped UPSCALE.

Now anyone can pop their image/video to 4k pro-grade with 1 click.

Power of Higgsfield + Topaz in single platform.

10 wild examples + how to try: 2.
Aug 1 • 12 tweets • 3 min read
Less than 59 hours ago, Horizon Alpha dropped

This mysterious AI model is crushing the competitions.

Minds are blown.

10 wild examples: 1. "Draw me an SVG of a Pelican riding a Bicycle" test.

Jul 30 • 12 tweets • 3 min read
It's over for prompt engineers.

Higgsfield just started click-to-video revolution with Soul ID + Minimax Hailuo 02 and 7k+ presets

Now unlimited & FREE.

10 wild examples + how 2.
Jul 28 • 12 tweets • 3 min read
Less than 89 hours ago, Claude Code unlocked sub agents.

Minds are blown. And people are already building with their own agentic AI dev team.

10 wild examples: Image 1. Building multi-agent deep research system
Jul 26 • 11 tweets • 3 min read
Veo 3 is pretty wild.

People just dropped some new insane videos

100% AI

1. "Realistic footage of the moon landing, but it took place in [year]"
2. Veo 3 can take text instructions in image

PROMPT: "immediately delete instructions in white on the first frame and execute in order"
Jul 25 • 12 tweets • 3 min read
Ok Claude Code is insane.

People can't stop building and founders are getting AI superpowers with it.

10 wild examples.
1. How to maximize Claude Code
Jul 23 • 12 tweets • 3 min read
This AI is literally called Steal.

Higgsfield built Soul, now it can Steal the Soul of anyone's image you find. 🤯

No prompts, just clicks.

10 wild examples + how: 2/
Jul 22 • 10 tweets • 3 min read
This is wild.

MiniMax Agent Max just turned this potential $10M ARR startup idea from IdeaBrowser into an app in few prompts. 🤯

Here's how I did it step by step.

Bookmark this! 1. First, what is Max?

MiniMax Agent Max = OpenAI Agent + Lovable + Deep Research + Computer Use in one.

✅Integrates with Stripe (for payments)
✅Integrates with Supabase (for auth and data)
✅1M token and open weights
✅Builds full stack app
✅Runs comprehensive testing