Strix

DeepSeek V4 Flash 0731 on 128 GB AMD Strix Halo: 82/92 quality, up to 32.7 tok/s decode, and 173 tok/s sparse prefill at 60K context.

This article has been edited and created by AI.Vulkan inference 12.9% faster: MMVQ tuning for AMD 780M/Strix in llama.cpp, Q4_0 ROCmFP4 models arriving one after anotherToday's LLM information for AMD 780M (RADV PHOENIX) includes several important discove

Apple

We gave Strix a domain. In 25 minutes, it found a live token with admin access to Baseten

AMD-Powered Lucebox runs the 284B DeepSeek V4 Flash model locally on AMD Ryzen AI MAX+ 395 with 128 GB unified memory, reaching up to 32 tok/s decode and roughly 250 tok/s sparse prefill.

Strix pointed itself at n8n

Strix found a one-click account takeover in Granola

On June 8, AMD opened pre-orders for a $3,999 box built around the exact chip I've run in production since the start of the year, marketed on full ROCm support, with one of its own demos running on the exact model my board can't load under ROCm.

The Bosgame M5's ROCm bug is board-specific, not chip-specific - so firmware is the obvious lever. I flashed Bosgame's newest official BIOS hoping to dodge it. It didn't work, and the negative narrows where the fault actually lives.

kmarble showed ROCm decode collapses 64% at full context on Strix Halo, and ROCm+MTP cures it. My board can't run ROCm. The Vulkan half reproduces the drop - but the MTP sweet spot from last week walks left at depth: by 76k, drafting too deep is slower th