Angry Tom (@AngryTomtweets) header
Angry Tom (@AngryTomtweets) profile picture

Consultant & AI educator | On a mission to build an empire with artificial intelligence

Joined October 2022
Dreamina just dropped a $0.29 deal for a full month of Basic... 1,575 credits. Enough for up to 10 10-second Seedance 2.0 videos. If the inference costs alone are anywhere near $10, how is Dreamina not losing money on this? 💀 Try it now: dreamina.capcut.com/pricing/… #DreaminaOnly29cents #Dreamina #DreaminaPartner #Seedance2
🤝 Paid partnership
14
2
31
4,799
wow... Google just took Genie 4 to another level. Genie 3 vs. Genie 4 side by side. no words.
This is Genie 4, our most realistic and interactive world model yet. 🌐 Explore dynamic environments in real time to see how the scene responds, sounds, and how the people in it move, react, and speak.
7
1
1
21
4,939
I am never deleting this app😭
13
4,913
That's what makes big action scenes interesting. You can think about the shot like a director: - where the camera starts - where it moves - what it follows
2
455
The impressive part isn't just the motion. It's the camera direction. The shot can move through the action, follow the fight, change perspective and keep the spatial logic intact. The model isn't losing the “where should the camera go next?” part of the scene.
2
4
676
AI video gets REALLY hard when the camera starts moving like a camera. Not a slow push-in. I mean fast tracking, sweeping movements, characters colliding, bodies flying across the frame. Kling 4.0 stays surprisingly stable when the whole scene goes BIG.
🤝 Paid partnership
8
3
1
26
4,997
I tried making a Tarkovsky-style film on Higgsfield for $4.20. Opus 5.5 studied his vision and storytelling through books and tutorials, then directed the whole thing. Wan 3.0 Prime generated it. this is unreal.
Higgsfield x Wan 3.0 Prime Switch between animation styles with faster generation at the best value. Available now at Higgsfield.
4
2
16
4,287
hold my beer...
You see how nobody is saying “I could make this with Higgsfield”?
98
243
33
1,509
113,140
I still had Viggle filed as the dance meme app. Then they shipped JST-2, one physics-native model for 3D characters, their motion and the scene they're in. You give it one image and get a rigged character back, then film yourself on your phone and it takes the motion from that. It's free today. The gallery is the fun part, you can walk around the worlds with WASD and screen-record whatever you find.
🤝 Paid partnership
Today we are releasing JST-2, the world's first full-stack physics-native world foundation model. Many world models learn in pixel/video space. We went back to first principles and built the entire model stack in 3D. Generate a 3D scene in seconds and expand it in real time. Create characters with motion built in, all through one general architecture.
3
1
17
3,233
Before subscribing, check the eligibility details. Your account must be registered in Japan, South Korea, the UK, France, Germany, Italy, Spain, Mexico, Brazil, the US, Canada, or Australia. The offer also requires your first-ever Dreamina payment. If you qualify, what kind of project would you use your month of Basic for? #AIcreator
3
495