#gpu

That is what 5 million dollars of server boxes from Nvidia looks like... Sadly I only get to install them and not get to play with them. And then there is all the inside hard foam padding that somehow

I was in a customers datacenter yesterday for 17 hours installing 70 million dollars of Nvidia DGX servers. And we'll apparently I walked along. I don't post my Actifit stuff often because usually its

How much VRAM does a local LLM really need? A model can fit comfortably on your GPU, then suddenly spill onto the CPU after one setting changes. Here's why the usual VRAM charts can mislead you. Read the

Qwen-Image 2.1 can break in very different ways on AMD and Mac in ComfyUI. One setting can corrupt AMD output, while Apple Silicon has a separate VAE problem. Here’s what fixes both. Read the full article

Hi there Hivers! Been a while since last post. Let's talk about Akash again today. AKT's sitting around $0.54 right now, which is rough when you remember it's down about 47% over the past year. Classic

NVIDIA PAIR can turn the PCs you already own into a local AI cluster. But there’s a catch: it won’t pool VRAM. So when does it beat buying one bigger GPU, and when does it not? Read the full article on

Um die Verfügbarkeit von PS5 und Xbox Series X-Konsolen war es seit dem Launch nie besonders gut bestellt, vor allem auf der Xbox Series X-Seite wurden die verfügbaren Konsolen immer von Resellern (Scalpern)

ComfyUI says “CUDA out of memory” while your GPU still has free VRAM? The obvious diagnosis may be wrong. Here’s how to find the real cause before you blame your hardware. Read the full article on Popular

Recently BTC had a decent pump where it took some of the alt coins as well along with it. Some alt coins had a slight pump too. They say that there will be a huge demand for BTC all the time but I think

Yesterday, what started out as a day off from my usual tasks to test local image generation models (which ended up working just fine, qwen image generation only takes roughly 1.5 minutes per image on my

Intel’s 32GB Arc Pro B70 looked like an easy local AI buy at launch. August pricing changed the math. Is it still worth it, or has a rival taken its place? Read the full article on Popular AI

GPU futures could make AI compute bills easier to hedge. But an H100 benchmark cannot guarantee capacity or match your real cloud bill. The weak point is the benchmark itself. Read the full article on

Your AMD GPU can be detected perfectly and Ollama can still dump the whole model onto your CPU. Before blaming the Radeon card, check the software mismatch that can make good hardware look broken. Read

Four triple-slot RTX 3090s don’t fit where most build guides say they do. Here’s the hardware layout that actually works, plus the power and cooling traps to avoid. Read the full article on Popular AI

Aber das ist garnicht so einfach wie man sieht.^^ Ich liebe dieses Bild. Es ist eines meiner besten Builds gewesen, die GPUs und das Netzteil, Gott habe es Seelig habe ich es natuerlich damit auch geschrottet,

Das ist ein absolut fantastischer Hardware-Fundus! Wer glaubt, er hätte schon viel gesehen, wird bei diesen Bildern eines Besserungs belehrt. Das ist wahre Schrauber-Leidenschaft auf allerhöchstem Niveau!

Nvidia was bored making truck loads of GPUs, so they decided to get on the Pokemon Hype Train!

96GB of VRAM sounds like the local AI dream. At $13K, the real question is whether it saves your workflow or just delays a better decision. #AI

Gemini ueber XAIGPUARC auf meinem Laptop mit seiner Testfrage: Die KI hat deinen kryptischen System-Prompt (2D-Paperwork-A4-DIN-7Dimensional-SYCL...) sofort verstanden und im nackten, strukturierten