Youtuber, I turn viral threads into videos with @vizplainer

San Francisco, CA
Joined March 2021
Made this video in 1 prompt 😉
Made this video in 5 prompts 😉
326
"99% is not 99% of the experience. It's zero percent of the experience." Thanks for @garrytan sharing!
1
281
Visual AI is shifting from generating pixels to generating code — so I used a visual AI tool to explain it. Fitting, right? Credit to @stuffyokodraws for the excellent article!
This quoted post is unavailable.
306
Some concepts hit harder when you see them moving. 🧠 Visualized this masterclass thread by @thedankoe . If you are a visual learner, this animation is for you. 👇
1
900
This thread completely rewired how I view AI workflows. 🧠 The concept of 'Onboarding' vs. 'Prompting' is the missing link for 99% of creators. I made a short video breaking down the system. Hope it can help @BawsaXBT Brick by brick. 🧱 👇
3
711
'3 Billion tokens later, and I’m done calling it 'vibe coding.' It’s a new programmable layer of abstraction. It’s the New Technical Class. Turned the story from @bentossell into a video.
3
3
22
6,715
Build something fully generated by AI. I don't know anything about motion design.
2
396
A85AN6 is a permanent invite code (according to the open ai discord) don't pay anyone for codes lol
1
3
871
Gemini 2.5 Flash Image killed it. REALLY GOOD consistency!
5
2,603
GPT-5 price is insane. How about claude?
1
594
A small test of FLUX knotext. It's good at detail preserving. But it still can only input one image to edit. The backbone is the same as FLUX fill.
2
3
796
For this nightmare garment, it's first time that we can see the Waistline when person wearing it.
1
1
595
BTW, the most impressive thing of FLUX knotext is the inferencing time. Hope the public weights release soon.
539
Black Magic An retweeted
BREAKING: 🇺🇸🇨🇳 United States cuts tariffs on Chinese goods from 145% to 30% for 90 days. China lowers tariffs on US from 125% to 10% for 90 days.
1,525
5,188
1,021
45,186
3,544,798
Next chapter starts
Embedding a scientific basis in pre-trained T2I models can enhance the realism and consistency of the results. Cool work in "Science-T2I: Addressing Scientific Illusions in Image Synthesis" jialuo-li.github.io/Science-…
1
670
GPT-4o downgrade its image generation model.
632
Black Magic An retweeted
👀⏳
118
138
73
1,228
142,804