RT tautologer: Fable attempts to communicate the gravity of this to me
RT tautologerFable attempts to communicate the gravity of this to melevent: hello there the jacobian conjecture is false thanx to my close friend akhil for asking…
RT tautologerFable attempts to communicate the gravity of this to melevent: hello there the jacobian conjecture is false thanx to my close friend akhil for asking…
no one is ready for if half of "mafs" is made up of trivial counterexamples that dumb monkeys have been dogshit at findingdoomslide: No one is…
I don't think mathematicians should worry much yetif the first part of Levent's bio were true, then for sureLisan al Gaib: drive-by humiliation of all mathematicianshe…
even as someone who is most definitely not a custom keyboards guy, i gotta admit this is pretty dang cool.
RT P Equity Research 📰Oh? 😂P Equity Research 📰: Huge order from SpaceX $SPCX for Foxconn to build over 13,000 racks of AI servers equipped with…
I wonder if @xenocosmography went to watchZhengxiao Han: Shanghai WAIC 2026
RT Bryan Cheong大道之行也天下为公Zephyr: @teortaxesTex Bro..
too much wordcelism around Dean's DecelgateI want estimates of 1) how much profits on inference do "The Labs" (don't fucking call them labs please, these are…
I feel like my timeline was right and now it is 3.5 months later.Assuming the Chinese government will still be okay with releasing open Mythos-class models…
RT Sakana AIRe “In Search of the Ingredients of Open-Endedness: Replicating Picbreeder with Large Vision-Language Models” won the Best Paper Award in the Complex Systems track…
One of the weird things about AI is some models just turn out to be much better than others and then the next model in line…
Fable, Sol Pro, Kimi K3: "write me a short but good poem using the Odyssey as a basis, think Tennyson or Cavafy"I think this is a…
about to write a bear scenario for DeepSeek and won't even get a few millions from Dario in return. pure love of truth
It's not so much that they underperform in real world useit's that they underperform in the higher weight class0.5-9B Qwens are monsters3.6-27B is great (though yes…
I'm no longer sure there will be a V4 GAYoussof Al Toukhi: Deepseek V4 GA is going to be the real monster. Not because of sheer…
didn't know that the Taiwanese got this shit tooI guess it doesn't really matter if you're a Good Asian or a Bad Asian…
Kimi doesn't have to be this slow btwthey're just really, really compute-constrained and so run it with extreme batch sizes to maximize throughputas a result their…
RT Alex CheemaI'll be sharing a first look at local dot ai with @huggingface on Tuesday.We benchmarked every model, quantization, harness, MTP setting, EVERYTHING, on every…
can we just say that Kimi didn't need to distill cyber?Ibrahim Dagher: A very interesting question will be how Kimi does on cyber, in particular stringing…
RT Vaibhav SisintyBaidu just open-sourced an OCR model that reads entire 40-page documents in one shot.It's called Unlimited-OCR. 3 billion parameters but only 500 million active…
RT cvbr.dsn🤹🚨вышла новая пипипупу, старая пупупипи забыта, more at 11しょー/AIの神: 今日のAI情勢
Riley has been doing some truly wonderful experiments with Fable. Also this is crazy.Riley Goodside: Crossword layout of all 1,009 distinct words of 12 or more…
> Today Chinese models account for the majority of token supply on OpenRouter, a huge increase over the past 18 months.stop making this argument. OpenRouter mix…
RT Sundar PichaiAn absolutely incredible performance by Spain to take home the World Cup! 🇪🇸🏆 So well-deserved, just a beautiful display of teamwork all tournament. Congratulations…