alby13 (@alby13) header
alby13 (@alby13) profile picture

Graphic Designer, Computer Tech, Video Game Designer. True Bitcoin Believer. X Posts are not investment advice. My data cannot be used for AI training.

USA
Joined December 2008
Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)Photo by alby13 (@alby13)
Apple autonomous agent
Now its getting interesting: Apple is building its own autonomous AI agent that writes code, browses the web, controls apps and navigates your devices, Bloomberg reports. A launch could come as soon as late next year.
2
1
53
Shubuya, Japan
Fable 5.2 built this exact replica of Shibuya in 12 hours. It even generated NPCs and simulated their behavior. EVERYTHING was done an an iPhone... using 8x NVIDIA L40S GPUs through Higgsfield GPU Harness. My Manhattan render took a week. This one took 2 hours. It’s fucking nuts.
19
I can solve the Riemann Hypothesis but I don't want to. 😂
History was made today. Humanity will look back on 2026 and think: this was the threshold we crossed, the moment the future changed for everyone, forever. People probably first sensed that something was shifting in late 2025 with Claude Code, when agentic AI became truly tangible for them. In the first half of 2026, things picked up speed. Model releases became more frequent, and everyone could feel the gaps between models and the improvements becoming more and more tangible. In the second half of 2026, however, there was a qualitative leap. It was no longer just the models getting better. It was the insights, the results, the outputs, the scientific solutions produced with AI that became more impressive, and today they have probably reached their absolute peak. OpenAI led the way and solved the Navier-Stokes problem, a Millennium Prize Problem that humans had failed to solve for a very, very long time. A few weeks later, they published results on more than 300 open mathematical problems, many of which had also gone unsolved for decades. The entire scientific world was stunned, frozen in shock. But today Anthropic went a step further. Levent Alpöge, working with Claude, has almost certainly proven the Riemann Hypothesis. This is, and it’s hard to put into words, incredible and unfathomable. Today, I would say, is the proof of superintelligence. One of the greatest mathematical problems, if not the greatest, unsolved for 167 years, the holy grail of mathematics and science, has fallen. And it follows, and this should be clear to everyone, that all other problems will fall too. We know that the development of AI models keeps moving forward relentlessly, and that this is just one step on the path to ever better models. And if the Riemann Hypothesis has already been proven today, there is simply no telling what even better models will solve. Dario Amodei said that AI will in all likelihood cure diseases, perhaps even all diseases. And although these are two different scientific disciplines, the significance of these breakthroughs shows that this is becoming more and more likely. 2026 will go down in human history as a turning point, the moment the singularity became tangible. And 2027 will be the year the singularity breaks into reality with full force and full impact. I am in shock and excitement and I don't know what to think. But deep inside if feel joy. This day is one of the best days ever for all of us.
1
40
Microsoft has released “Aion 1.0 Instruct,” a small LLM for Windows. It has 2.3B parameters. It was developed to summarize and rewrite text, understand documents, and answer questions on-device, and is designed to replace Phi-4-mini. Key specifications: -2.3B parameters, 32K context -Supports 8 languages, including Japanese -MIT license -Supports PyTorch, ONNX, and GGUF Microsoft plans to include a further optimized version of the model as standard in Windows OS by the end of January 2027. This means a model powering on-device AI in Windows has been made available for anyone to use.
Microsoftが、Windows向けの小型LLM「Aion 1.0 Instruct」を公開。 パラメータ数は2.3B。クラウドに接続せず、端末上で文章の要約、書き直し、文書の理解や質問応答を行うために開発された。 主な仕様は以下の通り。 2.3Bパラメータ、32Kコンテキスト 日本語を含む8言語に対応 MITライセンス PyTorch、ONNX、GGUFに対応 Microsoftは、さらに最適化したモデルを2027年1月末までにWindows OSへ標準搭載する予定。 WindowsのオンデバイスAIを支えるモデルが、誰でも使える形で公開されたことになる。 huggingface.co/microsoft/Aio…
1
75
Google launched Gemini 4 Neon. I've been testing it for less than a day inside Antigravity. My early impression is that it's really strong at coding and noticeably faster than Gemini 4 Carbon. Still testing and have a few more things to share, but here's something I built with it. Hope you enjoy it!
18
14
267
9,344
many fragments
A small light in an empty room. Move it, and everything moves away. 150,000 fragments, one invisible boundary. Physics in compute shaders, a custom depth of field, soft shadows and haze. All real-time, all in the browser. Built on WebGPU with @threejs r187 and TSL. 👉 the-light.m01-lab.workers.de…
54
AI can currently make GTA 3. It is getting close to 4. AI will be able to create GTA 5 and 6 in the future.
1
18
AI can currently make GTA 3. It is getting close to 4. AI will be able to create GTA 5 and 6 in the future.
🤖 Made with AI
1
1
4
290
Mechanical mechanics.
Damn, Google's Gemini 4 Neon is actually more insane than you can think I gave it ONE image of a robotic dog with a railgun and simply asked it to build it in 3D. No detailed instructions about physics or animation. And it ended up building an entire biomechanics engine using Three.js WebGPU. 240Hz rigid-body physics, 4-leg ground reaction forces, 3D inverse kinematics, realistic walk/trot/gallop cycles, dynamic spine bending, and even recoil-responsive head and tail physics. The amount of engineering it figured out is crazy. Also it felt super fast than Gemini 4 Carbon
2
233
Grok is insane RN - NSFW
🤖 Made with AI
2
103
it would be funny if Jensen Huang launches a competitor to NVIDIA.
1
3
242
city runner game - Gemini Neon
Gemini 4 Neon just gets it. Asked for a neon runner game in Shibuya playing as Pegman from @googlemaps street view. Simple prompt, /swarm /grill-me a few questions later and some nudges we have this. The demolab crew shipped 10 demos for yesterdays launch in record time. This model just cooks Pegman drops into the map at the scramble crossing to weave traffic, grind yamanote rails, and lights up neon signs on his way up to the finale at the Tokyo Google office on the crossing. Should we publish in the gallery?
1
143
will someone create a physics simulation and answer this? omg
Hay @grok , Can you tell me which one is strong here? One is connected with Boult and another one is connected with the help of Welding ( Heat) .
67
This looks really cool! ©️ duorug (Instagram)
18
234
13
3,361
266,243
it's fun working with realistic 3D models (FJ Cruiser)
2
81
a man doesn't need much to have a toy to play with
Real-time sand simulation & terramechanics in the browser (WebGPU + three.js) Study #3 Crashtests... Not GTA 6 – but the destruction-model is getting real. All physics, no canned animations. Runs in the browser. WIP @threejs @WebGL_WebGPU @WebGPU @80Level #destruction #threejs #WebGPU #gamedev #Simulation #physics #proceduralgeneration #desert
60
Steering wheels? Where we're going, we don't need steering wheels. A total Cyberpunk 2077 moment.
Check out the demo video of #BYD FCB Tai 9’s fold-away steering wheel.
59
I see fire, I see fire
《基于纯视觉识别的AI灭火系统》
2
65