will to code

Chinese labs that released models this week: - Moonshot: Kimi 2.6 - Alibaba: Qwen 3.6 27b - Xiaomi: Mimo V2.5-Pro - Ant Ling: Ling-2.6 1T - Tencent: Hy-3 preview 295b - Deepseek: V4 Pro/Flash And all of them are open source! imo this ai bet is the most important bet that the Chinese have ever made. If they even hesitated for a second they would've been left in the dust.

Once china gets the compute it's over

The whale is back 🔥

Just ignore the price lol Look at GPT 5.5 low on the chart. It's below glm 5.1 and kimi 2.6. Makes you wonder if compute is the real constraint. The Chinese models don't have low/high performance modes because they lack the scale? In that case Nvidia is def undervalued

They really cooked with this one

Gpt 5.5 just came out 🔥🔥🔥🔥

The Roman Empire: A Legacy of Failure Barely lasted. If it was like Egyptian Civilization it would still be thriving! imagine that What did the Egyptians have that the Romans didn't? > Mythology... their Pharaohs were divine This is clearly a much better form of governance but thanks to the "logic", "reason" and "philosophy" left to us by the Graeco-Romans we will never take it seriously to become a free-thinker was and still is the ideal but we must become free-spirits and mythologize once more Can you imagine a system lasting for 3000 years Longevity is all that matters If Rome had Egypt's longevity all the great discoveries of the 18th/19th century would have happened 1000+ years ago. CAN YOU IMAGINE WHERE WE WOULD BE NOW. That is why I view Rome as an absolute failure

kimi 2.6 just dropped looks promising next to claude... but don't look at gemini it is a shamelessly benchmaxed model

A service that loans out tokens with interest Most ROI than any other asset for both parties involved This asset, a TOKEN, has agency If I got a loan for a million dollars I can't ask it anything. I have to do everything myself. But if I got 100 million tokens for Claude.... I can ask Claude to do stuff for me We need tokens more than we need money do you understand it now Nvidea is undervalued The whole of AI is undervalued China built itself up by creating banks and using loans Now imagine the same process... creating data centers and getting Tokens as Debt. This is a glimpse into the abundance that is coming If you fail to pay your Token dept then your provider is incentivezed to keep giving you loans until you succeed. Because at the end of the day, ITS ONLY ELECTRICITY!!!

We owe ALL of the acceleration in the realm of technology to just a handful of optimistic based GIGA CHAD founders of these two Research Labs > BELL LABS > XEROX PARK

Search and you shall Discover!

Makes you wonder are iterations even worth it... What is the point in winning battles if your gonna lose the war... There may be some wisdom in the stagnation of Apple products. The empire iterating on different kinds of swords losses to the one that has discovered bows and catapults > Discovery beats iteration

*meta has joined the chat

The pentagon really fumbled this one Not good when AGI is your supply chain risk lol

ggs

It costs $25 input $125 output 💀

"AGI HAS BEEN ACHIEVED INTERNALLY"

New open source drop!!!! > 6x cheaper than opus

German word for library: Bibliotek 🗿

AI is wayyyyyy under hyped I was one of the people on twitter saying gpt-5 will be AGI in the early days of gpt-4: which is now considered an unusable model by any standards At this moment, we are the bottle neck: Skill Issue If you look at nations, or groups of people in general, the consensus always happens—or rather, policies always get enacted—after the fact. No action is taken during a decline in birthrate. For a better example: when the birth rate goes from 4 to 3, nobody sounds the alarm. It’s only when it reaches 1 or a negative number that policies start to get passed and the general public agrees that their birthrate is declining. In the same way nobody is aware of the insane performance leap from gpt-4 to gpt-5.4 > You just need to extrapolate a bit to understand the situation we are in You have to ignore the consensus of "ai slop" and try to push these models to any limit you can imagine them having