GLM 5.2 on Dual Strix Halo (256GB): Worth it?
Donato Capitella · 2 位 HN 用户推荐 · 2026–2026
I have two 128gb Strix Halos and have been extremely excited about Antirez's (Redis author) work on DS4, especially with 4bit quant using two machines: [link] Right now the speed isn't good for GLM 5.2, Deepseek V4 Flash speed is okay for me (actually reading the output) and quite usable. See kyuz0's great recent video here: [link]
— pixelpoet, 2026Currently, llama.cpp clusters don't support tensor parallelism, have a look at Donato Capitella's detailed report: [link] He also provides rocm toolboxes for Strix Halo: [link]
— mkesper, 2026