Trust no one.
Fine print: Not even yourself.
your brain literally lies to you constantly and you call it reality.
the human brain constructs what we consider to be reality. so you think you're seeing the world as it is, but you're not.
between what happens and what you see, there is always the brain. it doesn't observe. it predicts.
every second, it reconstructs the world from past experience, beliefs, and expectations. yet you mistake that reconstruction for "objective reality."
optical illusions simply make this mechanism visible. in real life, it works quietly; in money, relationships, decisions, fears, and repeating patterns.
but here's the part most people avoid: if perception is constructed, it can be rewritten. not positive thinking. not affirmations. but understanding how a thought become a filter of reality; how to exit autopilot mode.
when your thinking changes, reality doesn't "magically" improve.. it starts to look and feel different.
here's the simple truth underneath all of this: there's another way of seeing the world. there's another reality out there. it's right here, underneath the one you think you know. once you take action and you make different decisions, that's where everything begins to change.
3
Cognoscenti memory benchmarks are going to run for 16+ hours unless something breaks. 🔥
Time to watch an epic movie or something. Any recommendations?
2
1
46
- it took 20+ years for NVIDIA to hit the sweet spot.
- it took 20+ years for Spacex to become the behemoth
this was before industry was AI-led.
how many years would it take now for a hardware-pilled entrepreneur to become a household name?
clues anyone? 🙂
Replying to @DustinHuntwn
NVIDIA's edge is mostly software, not just silicon. CUDA, refined over 20 years with 11M+ developers and deep libraries (cuDNN, TensorRT), became the default for AI frameworks. Switching was slow and expensive. AMD and Intel GPUs are competitive on hardware (especially AMD on memory and cost), but the software lag kept most of the industry locked in. That gap is narrowing.
21
Does Google’s Gemini work for anyone?
It has benchmarks and a brand for sure, but the quality of response I get for some of the most simple and basic prompts is just plain <expletive>.
I do not think it can compete with the likes of Anthropic or OpenAI at all.
Heck, it doesn’t hold a candle to Qwen even, which is much more compact and open weights. 🙊
17
Marvin Danig retweeted
A supercomputing lab on your desk. ⚡
New at #SC26: the NVIDIA DGX Spark HPC Developer Kit, built for researchers, academics and developers.
🟢 Connect up to 4x DGX Spark systems in a ring over NVIDIA ConnectX-7
🟢 No network switch needed
🟢 Up to 512GB of memory
Marvin Danig retweeted
Your customers are your accelerator.
Future of abundance = abundance of human intelligence, mainly.
Everything else requires mobilization.
17
Oh, I love this new Qwen4 release. The numbers are just insane.
qwen.ai/blog?id=qwen4-max
Like I've said before, open weight is the new frontier… and it's like linux going against bozos all over again.
I'm excited about our future.
31
even a low or medium risk vulnerability may be a problem in face of an unrelenting "agentic" hacker.
15
now this is a Chinese model i can easily get behind. 🐒
Qwen4 is coming.
Deeper knowledge. Richer world understanding. Stronger coding and agentic capabilities.
Built for real-world tasks — from complex coding and knowledge work to front-end builds and professional software workflows.
Qwen4-Max is now in preview on Alibaba’s Token Plan. You can also get early access to the new model on QwenWork, Qoder, Qwen APP, and QwenStudio.
Put it to the test on a real problem. Show us what you build. 👀
Token Plan: qwencloud.com/pricing/token-…
9
















