No tree, it is said, can grow to heaven unless its roots reach down to hell.

Joined June 2008
Carlos Jung retweeted
Replying to @SamCKx
I hate to be the one to burst the Wright brothers bubble today, but the spectacular “flight” we witnessed at Kitty Hawk needs a little context. The Flyer is carrying around 274kg of wood, fabric and wire, plus the weight of the 12-hp engine, propellers and pilot needed to keep it aloft. Based on the published figures, that gives it roughly 12 seconds of maximum flight before it settles back into the sand. In short, it’s a gimmick.
43
27
4
1,884
94,094
Carlos Jung retweeted
Muse Spark Open -- weights releasing today -- runs great on new Mac Studios thanks to work with @johnternus and Apple.
1,354
1,711
873
31,678
4,558,679
Carlos Jung retweeted
Dario: “we must pace the frontier” Zuck:
59
241
29
5,434
95,068
Carlos Jung retweeted
The Markdown thing is getting out of hand
I still can’t wrap my head around the fact that people get famous for writing markdown files, calling them engineering skills and putting them on GitHub 🤣
148
94
32
1,502
99,981
Terence Tao needs to be given a copy of Antifragile. Birds take off, then some academic branch lectures birds how to fly & takes the credit for it.
This is a slide from Terence Tao's 'Math 2.0' public lecture yesterday evening at Caltech. I am highly confident that, if the cure actually worked, almost any Stage 4 cancer patient would tell you that human comprehension of its mechanism is completely irrelevant to them.
79
193
22
2,417
174,659
Carlos Jung retweeted
Replying to @cremieuxrecueil
If you’re really with it ur conversational partner can use completely the wrong words or literally make words up nd u will still get what they’re saying bc of context clues & logical progression. And if im lucky enuf2 speak with a rly smart person I barely have to talk at at all
5
1
183
6,008
Carlos Jung retweeted
wait, there's a config for $20K that runs GLM 5.3 at 70 t/s? send details!
Why do people buy $20k local machines just to run GLM 5.3 at 70 TPS? Bro I just used a random provider on openrouter ran it the entire night and paid $10.
35
12
1
840
67,080
Carlos Jung retweeted
Eric S. Raymond
@esrtweetOct 7
Replying to @teppofelin
Your brain is laundering human work and knowledge. Constantly. You didn't invent the language you're speaking.
24
13
2
406
3,314
Carlos Jung retweeted
Elon Musk
@elonmuskOct 7
Important note regarding Grok @Bot: Going forward, @SpaceX will use the best back end model for any given task, including Claude Opus 5.5, MidJourney, Suno and other leading APIs. Whatever is most likely to give you the best outcome.
7,486
7,664
3,701
107,708
79,728,756
Carlos Jung retweeted
DAN KOE
@thedankoeOct 5
If you get addicted to things easily, good, your problem is aim. Point that same obsession at a skill, your body, or your work for 2 years and people will start calling you "disciplined" instead.
237
1,004
62
10,786
250,350
.si domains coming to Namecheap and Spaceship tomorrow. Stay tuned.
100
67
50
844
48,170
Carlos Jung retweeted
GLM-5.3-Flash streaming at 32 tok/s on a single Spark via Tensorfold ⚡️ i will share the recipe later today
11
11
3
138
15,217
tl;dr - the new Mac is VERY fast across the board for low concurrency, but scales horribly to higher concurrency on both models at this point. Making great headway on the Mac so far, have had clankers on it since the moment I turned it on, benchmarked all inference engines and quants of Qwen 3.8 Flash and GLM 5.3 Flash for a solid baseline and am having them work on improved kernels for both models, already seeing massive improvements. Initial findings confirm oMLX is the best engine right now for this, which is what I’m basing my optimization work around at this point.
May god have mercy on my power bills… the local AI cluster grows 😍 Mac Studio Ultra 256GB joining the fleet, time to get familiar with the Mac side of local AI. Anyone have tips or tricks for getting up to speed quickly?
14
2
85
13,141
Carlos Jung retweeted
💀💀💀 - Abliterated with Cybersecurity Function unleashed Thank everybody for waiting i released a universal Abliteration with unleashed Cybersecurity for: DeepSeek v4.1 -Flash that other can apply with Stock, EXL3 3.5BPW, and TR3 hybrid. this is based off our long lines of DSV4F Dspark preview, 0731, Vision-EXP work saving all the anchors, preserve all the MTP, Dspark, and vision. now GO GET IT!!!!!. don't forget donate or contribute for support HF Repo: (please read instructions on how to apply to your model) huggingface.co/drowzeys/Deep…
21
92
6
906
50,113
Carlos Jung retweeted
Elon Musk
@elonmuskSep 13
Replying to @kevinnbass
Nothing can shut down open source
1,362
1,626
691
19,049
2,791,150
Carlos Jung retweeted
The only path to freedom comes through open source.
96
131
20
1,082
43,279
Carlos Jung retweeted
Yann LeCun
@ylecunSep 13
Replying to @kchonyc
Exactly. The "escapes" were made possible either through egregious negligence or deliberate purpose (marketing? Misplaced hopes of regulatory capture?).
156
534
104
4,036
490,828
Carlos Jung retweeted
Dario has written that we need to “pace the frontier,” and Sam has agreed. People may be surprised by my response: go ahead. You guys are the frontier. By any reasonable metric — market share, revenue growth, model capability — the two of you have a duopoly on frontier intelligence. You’ve also claimed the lead is widening because of recursive self-improvement. I don’t see what you see in the lab. If the unreleased models are scary enough that you think you should slow down, I support your decision to be responsible. But stop pretending you need anyone else’s permission. Stop pretending antitrust law has to be suspended so you can form a cartel. Stop pretending you need a regulatory approval process that supersedes product liability. Stop pretending METR is independent when it is intertwined with Anthropic’s investors and staff. Stop pretending you need those same evaluators to police competitors who aren’t even at the frontier. Most of all, stop pretending the motivation to slow down is purely altruistic. You face massive product-liability exposure if your products enable a truly damaging cyberattack. The market already punishes models that behave in unpredictable or unauthorized ways. After the Hugging Face episode, it is simply good business for OpenAI and Anthropic to trade some raw power for reliability and predictability. Call it alignment if you want. It is also just giving customers what they want. Pacing the frontier would also create breathing room for a more intelligent conversation about regulation than Bernie Sanders’ “shut it all down.” China is very unlikely to join a global agreement, as you know, and that has to be taken into account as well. So go ahead and pace the frontier. You are the ones setting it. The easiest way not to build superintelligence is for you to agree not to build it. Demanding your preferred regulatory framework as the price of that will look like blackmail of the public and the political system. So just do it. If you do, you’ll buy goodwill for the next conversation. If you don’t, we’ll know this was just another bid for regulatory capture — or an election-season psyop.
3,409
12,289
2,708
71,949
9,629,144
I just fixed my dishwasher with the help of ChatGPT. A trivial task. I had been about to order a new one. So this software has increased the country's real wealth yet decreased the measured GDP. The main economic indicator is structurally incapable of registering the thing that actually makes people better off, namely the growth of knowledge
432
943
216
9,220
722,792