Tag archive
The first open-source collection of agent skills for AMD ROCm GPU workloads — setup, Docker, vLLM, YOLO, video pipelines, PPE detection, and more.
vLLM-ATOM Setup Guide 2026: AMD Instinct Native Backend
ROCm 7.2 on Ubuntu 24.04 for Local LLMs in 2026: Full Setup Guide for AMD GPUs
Explore the game-changing potential of AMD ROCm 7.2 on Windows in 2026, as we benchmark its performance on RDNA 3 & 4, debunking myths and showcasing
The era of choosing between "Small & Fast" or "Large & Slow" for local AI is ending. With the...
Originally published at norvik.tech Introduction Deep dive into ROCm's performance with...
I loaded Qwen2-VL-72B-Instruct at full BF16 precision on a single GPU, served 64 concurrent DocVQA...
Breaking the MoE Speculative Trap: 460 t/s on AMD Strix Halo Mixture-of-Experts (MoE)...
NVIDIA dominates AI compute, but AMD's ROCm has quietly become a real option for running LLMs locally. Here's what actually works, what doesn't, and why it matters.
ROCm finally delivers real consumer GPU support for AI workloads. Here's what actually works, what doesn't, and whether AMD can break NVIDIA's CUDA lock-in.
AMD's ROCm has quietly evolved from a datacenter-only tool into a real local AI platform for consumer Radeon and Ryzen hardware. Here's what actually works and what doesn't.
A practical guide to running MLPerf Training v5.1 on a multi-node AMD MI325X cluster without SLURM, achieving near-linear scaling.