From google research earlier today on X. … Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI It’s possible that it makes its much easier to run models locally..

Overnight block 940,073 was mined. This block includes the 20,000,000th bitcoin. Only 1,000,000 to go. True provable scarcity.

Sounds like a good way to keep the population poor…. The lengths that politicians will go to in order to extract value from their constituents (slaves. ..) continually amazes me. View quoted note →

@Kenshin 🥷 interesting take … View quoted note →

We run the network!!!

I am 100% certain that @Kenshin 🥷 and I did not talk about Davos at all. Because we are out building things and talking about Bitcoin instead of worrying about what a group of people who have no interest in you have to say. Satoshi’s Plebs episode 241 is out by the way!! Grab it on your favorite podcast app and give it a listen.

One of the few heros I have in my life!! View quoted note →