TECHNOLOGY

–

25 March 2026 22h23

Move to Linux and try LMStudio and then Exl2/Exl3

→ View original post on X — @ronald_vanloon

Human-Scale Bipedal Humanoid Robot for Real-World Tasks

By

@ronald_vanloon

–

25 March 2026 22h19

Human-Scale Bipedal Humanoid #Robot Designed for Real-World Tasks
via @ZappyZappy7
#Robotics #RPA #ArtificialIntelligence #Innovation #Technology pic.twitter.com/j5P47UBkpQ
— Ronald van Loon (@Ronald_vanLoon) 25 mars 2026

Human-Scale Bipedal Humanoid #Robot Designed for Real-World Tasks
via @ZappyZappy7 #Robotics #RPA #ArtificialIntelligence #Innovation #Technology

→ View original post on X — @maximelabonne, 2026-03-25 21:10 UTC

24B Model Running in Browser with WebGPU Technology

By

@maximelabonne

–

25 March 2026 22h10

24B model running in your browser! https://t.co/dorx4Ux2x7
— Maxime Labonne (@maximelabonne) 25 mars 2026

24B model running in your browser! Xenova (@xenovacom) WebGPU is INSANE! 🤯 Here's a 24B parameter model running locally in a web browser, at a blazing ~50 tokens/second on my M4 Max. ⚡️ It's the largest model we've ever run with Transformers.js… and we're not stopping here. Big announcement soon. — https://nitter.net/xenovacom/status/2036908326462665211#m

Abstractions hinder performance, bare metal is optimal

By

–

25 March 2026 22h01

Abstractions are not great for performance Bare metal is the way to go imho

→ View original post on X — @levelsio

OpenAI Debuts Major ChatGPT Ads Updates, Including Self-Serve Manager and CPC Bidding

By

@levelsio

–

25 March 2026 21h57

Most interesting image from this article Everyone changed from classic incadescent lamps to modern LEDs But they don't provide any infrared light Even the warm LEDs you buy barely do We really need incadescent lights back but the EU made those illegal, or we need to buy those

Warning: Do Not Use Windows for Local LLM Deployments

By

–

25 March 2026 21h44

I was assuming it’s common knowledge at this point not to use Windows for local LLMs But just in case: DO NOT USE Windows for local LLMs

Hardware Alone Isn’t Enough: Software Stack is Crucial for Adoption

By

–

25 March 2026 21h28

It’s wild to me that folks still don’t understand this A piece of hardware doesn’t matter if the software stack isn’t there Good luck waiting 5 years for that Intel GPU to be widely supported Even the Blackwell kernels aren’t mature yet and it’s already widely adopted lol

→ View original post on X — @askalphaxiv

NOBLE: Nonlinear Low-Rank Branches Accelerate Transformers

By

@askalphaxiv

–

25 March 2026 21h24

"NOBLE: Accelerating Transformers with Nonlinear Low-Rank Branches" This paper shows that if you add a tiny permanent nonlinear low rank branch (with a cosine bottleneck) to each Transformer linear layer, you can teach the model hard-to-fit details much more efficiently. This