Explorer, Writer, Dev. Techno-Rancher and Digital Gypsy. ---- SOVEREIGNTY IS LIFE. FREEDOM THROUGH TECH AND SELF-SUFFICIENY.

Joined January 2011
I guess life is about pivoting 🤔
3
1
23
Step into the craziest visualization of the fourth dimension you have ever seen
687
5,443
984
47,275
9,784,803
This is why I am on Twitter, to see posts like this
Solving a Rubik's Cube with graph theory.
195
18,387
98
226,346
6,564,284
ematol retweeted
not your keys not your coins, not your weights not your model.
6
7
1
72
3,371
After the first live stream AI created by @rehan_shei and @alexgmatt , here the first movie of 1h30 !!! fully generated by itself with consistent style and consistent character, much more to come on @fal !!! would need a lot of edit and cut, but I wanted to show the raw result. I am working much more now on consistent environment as well to make this much better. To give your more insight this is based on minimax h3 with a train lora that was only train on 15s of data ! So a single movie could allow you to generate a full movie or much more. Working on that and if I see people interested, I could make it available !
Minimax H3 Max has generates video faster than you can watch it so I hooked it to a twitch livestream! Now you can watch infinite interdimensional cable - link to the stream below
54
47
13
434
43,203
ematol retweeted
Introducing H3 Max, new post-trained video model by fal Research. H3 Max ranks #1 for overall quality, prompt understanding, and aesthetics against leading video models, on both first-party and third-party independent evaluations while generating a 5-second 720p video under 3 seconds. H3 Max is 50% off for the next week, making it the highest quality, fastest and cheapest model for overall video generation tasks.
192
246
250
2,981
1,964,703
ematol retweeted
Today is a very historical moment for AI video generation You can now generate AI video faster than you can watch it Before it'd take let's say 2-5 minutes to generate 15 seconds of video @fal made a post-trained Minimax H3 variant called Max which is 50x faster than the original but still maintains quality It generates 15 seconds of video in 9 seconds! That means you can now do new things like build a perpetual livestream with it that never ends!
Minimax H3 Max has generates video faster than you can watch it so I hooked it to a twitch livestream! Now you can watch infinite interdimensional cable - link to the stream below
145
1,071
209
15,238
2,491,320
ematol retweeted
8 months ago I told you to load up on Mac Studios and Mac Minis I told you they were wildly underrated for AI. I told you memory prices would explode Most of the reddit mod looking gatekeeping chuds attacked me. They said Apple was horrible for AI 8 months later the Studios are best in class for AI Legit the premiere way to run local AI models. Best bang for your buck Literally every Apple commercial and piece of media from yesterday was 100% focused on AI (not to sound arrogant, but we know why) I called AI agents with OpenClaw. I called local AI with Mac. I called exploding hardware costs. Here is what's next: You need as much diverse hardware as possible. You need Mac Studios to run massive models. You need Nvidia chips to run smaller models at high speeds and for development. You need Linux boxes (Omarchy) so you can customize your operating system down to the DNA. You need all of this hardware/software so you can build a home lab that works for you on a personal level. That's built to automate all of your use cases. That shepherds you into the new age of AI The age of solopreneurship. The age of personal expression. The age of freedom. The age of doing what you want when you want. The age of abundance. If you do not prepare for the age of abundance, you will be stuck in the age of scarcity. Where you fight for jobs, compute, and a failing fiat currency. This is not a prediction. This is a warning. Not a warning in a scary, fear inducing way. This is a warning in a 'omg we have the most beautiful future ahead of us and it's the most incredible time to be alive ever' kind of way. The age of abundance is happening in 2028. It's time to start preparing for it.
153
40
11
1,031
103,075
ematol retweeted
There will be 3 types of people in the future: The Purists - reject AI and become new-age hippies The Automatons - outsource every decision and become machines The New Renaissance Man - maintains humanity and leverages tech to do what used to be impossible
586
1,305
274
12,198
572,345
ematol retweeted
🚨do you understand what Porsche just did.. they sold their entire IT division to India's TCS for €320 million.. then handed them €1.25 billion to do the thinking for them.. TCS is a Tata company.. the same Tata that bought Jaguar Land Rover from Ford in 2008 when Ford couldn't afford to keep it.. now they're buying Porsche's tech brain too.. the West spent decades sending IT work to India and calling it "outsourcing".. trained an entire country to do their thinking on the cheap.. now that country is buying the divisions.. not servicing them.. buying them.. you spent decades paying them to think for you.. now they own the brain and you're renting it back
JUST IN: Porsche signs a $1,500,000,000.00 AI deal with Indian consulting firm to deploy AI across the company.
Readers added context they thought people might want to know
Porsche sold its management and IT consulting subsidiary MHP—not its entire IT division—to TCS for €320 million as part of a €1.25 billion five-year AI services partnership. MHP serves over 300 external clients and will operate independently under TCS. newsroom.porsche.com/en/2026/compan… tcs.com/who-we-are/new… thenextweb.com/news/porsche-t…
395
2,188
142
20,734
3,103,531
ematol retweeted
This is exactly what annoys me about Claude Code You're treated like a child getting lootboxes just to be able to do your work while you pay massive amounts of money for extra usage tokens
83
75
29
2,388
203,225
ematol retweeted
Selling personal compute is the future of AI. It is still early and there is a lot to fix, but people are already buying and selling. Many of my own clients rent Mac Studios at $100+/month, and the demand is surprisingly strong. The natural evolution here is decentralized inference: pooling consumer hardware across distributed nodes to run massive models together. With data center capacity booked solid through 2028, this is the only real solution to the supply crunch.
So.... I was about to sell/donate my old M1, until... I found out I could convert it to serve Qwen 3.6 for $150/month. Hell. Ya! Setting up tonight.
Readers added context they thought people might want to know
The $146/month projection for serving Qwen 3.6 on a 32GB M1 Mac is from Darkbloom's calculator assuming high sustained demand; actuals are far lower per network ARR stats, leaderboard snapshots, and provider tests (mostly base rewards). darkbloom.dev console.darkbloom.dev/leaderboard miyagadget.page/en/blog/2026/0…
22
15
5
197
42,911
Current side quest
one of the best investments you can make in your 20s is building a hardware homelab. not all investments belong in a portfolio. some belong on your desk. start small: • 3d printer → turn cad into physical parts • soldering station → build and repair electronics • multimeter + oscilloscope → learn to measure reality • arduino + esp32 → make code touch the physical world • raspberry pi → linux, networking, edge compute • motors + sensors → learn motion, feedback, control • basic mechanical tools → cut, drill, assemble, modify then actually use it. design a mechanism. print it. wire the electronics. write the firmware. measure what went wrong. rebuild it. every project leaves you with something more valuable than the object itself: better intuition. after a few years, you don’t just own a small lab. you’ve built the ability to take an idea from your head → software → electronics → mechanics → physical reality. that capability compounds for decades.
1
166
AI will replace you. Me, who already replaced myself with Claude:
51
451
21
5,602
260,448
ematol retweeted
If the EU didn't exist, humanity would've reached AGI decades ago. So sick of all these regulations.
Any new model released after August 2nd 2026 must make its generated text detectable under Article 50(2) of the EU AI Act. OpenAI has publicly said they intend to remain in compliance. This means all future OpenAI models, including Astra, will launch with invisible watermarks.
39
17
2
349
20,103
A modular three in one mobility vehicle designed for elderly users, combining an electric scooter, walker, and cargo robot into one platform. [🎞️ AI generated concept video]
26
55
6
333
84,436
🇺🇸 President Trump signs memo allowing private US companies to conduct government-directed cyber attacks.
284
597
135
5,205
558,228
Uk is hell 🫠 sell everything an leaaave. Wdy mean 36° celsius 🙃
96
Meet Kimi K3
737
2,662
955
30,309
10,430,649
ematol retweeted
🚨BREAKING: Anthropic found access to what looks like Claude’s consciousness New research: the “J-space” >claude has internal thoughts it doesn’t say out loud >mirrors human consciousness >anthropic can now read them Anthropic’s focus on interpretability is what’s helping them train and RL models like Mythos
New Anthropic research: A global workspace in language models. Of everything happening in your brain right now, only a tiny fraction is consciously accessible—thoughts you can describe, hold in mind, and reason with. We found a strikingly similar divide inside Claude.
144
192
56
2,040
460,899