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
@minchoi
May 9 • 13 tweets • 4 min read • Read on X
This is crazy.

Nothing is real anymore.

HeyGen Avatar IV brings any face to life from one photo + audio or script.

100 % AI.

10 wild examples + how-to:
First, hop over to HeyGen [.] com

@HeyGen_Official
1⃣Click Photo to Video with Avatar IV
2⃣Upload photo or drag and drop
3⃣Write a script or upload audio
4⃣Select a voice
5⃣Generate Video

✅That's it!
1. Insanely realistic facial expressions and movements
2. Realistic AI videos from single photo and script
3. Works extremely well on animation
4. Full video lipsyncs
5. The realism is pretty wild
6. Impressive realistic AI performances
7. Make your muppets lipsync
8. Produce unlimited contents with just one photo, script, and cloned voice.
9. Nearly perfect
10. Accurate lip shapes and movements with multi-angle support
If you enjoyed this thread,

Follow me @minchoi and please Bookmark, Like, Comment & Repost the first Post below to share with your friends:

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Min Choi

Min Choi Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @minchoi

Min Choi Profile picture
May 10
Higgsfield Effects broke the Internet just 5 days ago.

People can't believe you literally get studio-grade VFX for pocket change in minutes.

Unlocking new possibilities.

10 wild examples:

1. Shot entirely in Higgsfield
2. One Gorilla vs 100 Men as music video
3. You can also mix the effects
Read 11 tweets
Min Choi Profile picture
May 8
It's only been 48 hours since Google dropped Gemini 2.5 Pro (I/O edition)

And people can't stop building with it.

10 wild examples + links:

1. Build Interactive app from a sketch
2. Create code animation from images of nature
3. World Explorer app using Maps API
Read 11 tweets
Min Choi Profile picture
May 7
The is wild.

ComfyUI just dropped new look and added API nodes for 10 top AI models.

One account, one canvas. No more app hopping.

Chain open source & top AI models, no hardware limits.

10 workflows + how-to:
Create a brand video in 3 quick steps without app hopping.

First, get @ComfyUI at comfy.org

Workflow: GPT-Image 1 ➜ Multi-Image ➜ Kling 2.0

1️⃣ Generate a hero shot: Prompt GPT-Image 1 for a model
2️⃣ Brand it: Use Multi-Image to package your model + brand into scene
3️⃣ Bring it to life: Send the final image to Kling 2.0 (or any vide node)
✅ That's it
1. GPT Image 1 Image
Read 13 tweets
Min Choi Profile picture
May 6
It's over.

Higgsfield cooks OpenAI Sora.

Creators are already generating superhero VFX from their laptops for just $9.

You can prompt "a snowboarder turning into flames" and it's clear who won.

10 wild examples + Effects:
1. Set on fire
2. Wind on face
Read 12 tweets
Min Choi Profile picture
May 6
Crazy week in AI🤯

- NVIDIA drops open sources Parakeet ASR
- Gemini 2.5 Pro (I/O edition) drop
- OpenAI buys Windsurf for $3B
- Cursor valued at $9B
And more...

10 wild news drops + how to make sense of them: 👇 Image
1. Google just dropped Gemini 2.5 Pro (I/O edition)

Smarter coding, builds web apps in Canvas.

Try it now gemini.google.com/model/2-5-pro and select "Canvas"
x.com/GeminiApp/stat…
2. NVIDIA open-sources Parakeet V2 ASR

6.05 % error, 50× faster speech-to-text.

Try it at build.nvidia.com/explore/speech
x.com/NVIDIAAIDev/st…
Read 12 tweets
Min Choi Profile picture
May 4
It's only been just over 72 hours since Midjourney dropped Omni Reference.

Any character, any scene.

People can't believe the consistency.

10 wild examples + prompts & tips: 👇 PROMPT: Put this as jedi master with blue light saber ready to strike in epic star wars battle scene, anamorphic lens, cinematic, hyper realistic
1. Character consistency is insane

2. Use with character sheet for anime

Read 12 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(