i know way too much, yet i know nothing
somewhere on planet earth
- Tweets 12,692
- Following 333
- Followers 2,104
- Likes 45,689
i can't be arsed to find it, but there's clear proof i called this out a month ago.
anyone who thinks any human female who looks even remotely like this is into local AI, let alone a leading face of local AI, is delusional
Local AI is a boys' game, just like gaming is, and was until the girls were forced to be allowed into the space.
this is basic biology. if you're one of the dudes who dreamed about marrying this Mia chick, you need to uninstall all software and seek professional help, and maybe go read about human evolution or something else basic.
thank you.
They built a 37K cult using a fake avatar to hijack the algorithm. Code never lies and who is actually behind it is revealed inside.
Model credit: @ViC305 & @Blackfrost_ai (not participants).
Read the receipts behind Mia’s AI Lab 👇
3
99
never give up, never surrender
broke 100wpm for the first time. this is what tailscode does to a mf
1
6
336
AMD RX 7900 XTX is a pretty good option for gaming/LLM use. It has the same bandwidth as an RTX 5080, and ROCm is getting closer to Nvidia's stack. LLMs are getting good and small enough to fit on a card with 24GB. Plenty of stock on the second-hand market. I'll probably put mine in a dedicated LLM server box when Intel's new CPU comes out. Too often, I'm bottlenecked because I wanna do local inference and play games.
1
108
If you're insecure about sunlightless virgins hazing you, you need to realize that agentic tools allow you to remove them from the equation entirely. Arch Linux has no barriers between you, your LLM, and the entire system.
It is time, Jared.
Because the setup of Linux has been treated like a hazing ritual of config and driver hell for decades?
2
150
Consumer and enterprise robotics will see a massive increase in demand by the end of 2027. This hardware requires even faster memory than is available today.
Gaming GPUs will not become cheaper. They might disappear entirely, or we'll plateau at the current level, which is fine. The problem has never been the graphical fidelity. It's always been the gameplay loop.
Agentic tools will allow game devs to optimize their software by an order of magnitude, and everything will run better. Nvidia's software engineers will also work on this. It's gonna be fine. wagmi.
en.gamegpu.com/news/zhelezo/…
The upcoming GeForce RTX 6090 graphics card based on the Rubin architecture is unlikely to debut before late 2027 or 2028. Analysts cite high GDDR7 memory costs and wafer manufacturing allocations favoring data centers as the primary reasons behind this timeline revision.
Server accelerators consistently maintain gross margins exceeding 90%, which naturally pushes enterprise orders ahead of consumer demand. The desktop GeForce lineup is no longer the main revenue driver, serving primarily to protect overall brand status and enthusiast market visibility.
Decoupling consumer and compute dies onto older process nodes remains economically inefficient. Barring a dramatic enterprise slowdown, the flagship GPU will face extended delays and launch strictly as a premium ultra-enthusiast product.
#nvidia #geforcertx #rtx6090 #hardware #gpu #pcgaming #tech
1
5
525