Leave your feedback Share Copy URL https://mypepack.gophersport.com/video/?vid=dLP5VaWHsqh Email Facebook Twitter LinkedIn Pinterest Tumblr Share on Facebook Share on Twitter AMD’s CEO Destroyed NVIDIAs Most Expensive Supercomputers With a $1,500 Lunch-Box PC! [mNwLcyZnOMB] Health Updated on August 07, 2026 EDT — Published on August 07, 2026 EDT AMD may have just exposed the biggest problem with Nvidia’s personal AI supercomputer strategy: price. While Nvidia’s DGX Spark costs around $4,699, AMD’s Ryzen AI Max+ 395 mini PCs are bringing serious local AI power to a much lower price range. In this video, we break down AMD’s Ryzen AI Max+ 395, codenamed Strix Halo, and why this chip is becoming one of the most exciting local AI hardware stories of the year. With 16 Zen 5 cores, Radeon 8060S graphics, an XDNA 2 NPU, and up to 128GB of LPDDR5X unified memory, Strix Halo can run large AI models locally without needing a massive GPU tower or cloud subscription. We also compare AMD’s Strix Halo systems against Nvidia’s DGX Spark, looking at local LLM performance, memory bandwidth, and llama.cpp benchmarks, CUDA vs ROCm, Windows support, and why AMD’s cheaper AI mini PCs could be perfect for developers, students, hobbyists, and small studios. Could AMD’s $1,500 AI lunchbox make Nvidia’s $4,699 AI box look overpriced? #AMD #RyzenAI #StrixHalo #Nvidia #DGXSpark #LocalAI #AIHardware aqQSvizjQhi QNBd9GXiMEk bsY7xkuQ6Wz zdiN3GmJzna Zg0g8yP0bw8 nLxD44op8ud
AMD may have just exposed the biggest problem with Nvidia’s personal AI supercomputer strategy: price. While Nvidia’s DGX Spark costs around $4,699, AMD’s Ryzen AI Max+ 395 mini PCs are bringing serious local AI power to a much lower price range. In this video, we break down AMD’s Ryzen AI Max+ 395, codenamed Strix Halo, and why this chip is becoming one of the most exciting local AI hardware stories of the year. With 16 Zen 5 cores, Radeon 8060S graphics, an XDNA 2 NPU, and up to 128GB of LPDDR5X unified memory, Strix Halo can run large AI models locally without needing a massive GPU tower or cloud subscription. We also compare AMD’s Strix Halo systems against Nvidia’s DGX Spark, looking at local LLM performance, memory bandwidth, and llama.cpp benchmarks, CUDA vs ROCm, Windows support, and why AMD’s cheaper AI mini PCs could be perfect for developers, students, hobbyists, and small studios. Could AMD’s $1,500 AI lunchbox make Nvidia’s $4,699 AI box look overpriced? #AMD #RyzenAI #StrixHalo #Nvidia #DGXSpark #LocalAI #AIHardware aqQSvizjQhi QNBd9GXiMEk bsY7xkuQ6Wz zdiN3GmJzna Zg0g8yP0bw8 nLxD44op8ud