EVO-X2

The GMKtec EVO-X2 with A.I large model feature edge advantage and the Minisforum G7 TI for raw Power.

Thank me later.😎 Strix Halo 128gb Laptop on sale, killer deal. https://www.microcenter.com/product/702348/hp-zbook-ultra-g1a-oled-14-mobile-workstation-pc-meteor-silver Minisforum Strix Halo 128gb mini PC https://www.microcenter.com/product/702314/minisforum-ms-s1-max-next-gen-ai-workstation Better

FN60G with a 14700KF and 4090M GPU or Lenovo P3 Ultra with a A5500 GPU. Honerable mention to the GMKtec Evo X2 and Bosgame M5 for having the most powerful CPU/iGPU combination. And honerable mention to ITX builds with a 9800X3D and 5090 GPU. 2025 General Mini PC Guide

Agree that it’s difficult to recommend a mini PC with that budget when you could build a sub 5L SFF PC with a lot faster GPU for gaming than an AI Max mini PC. But if you absolutely needed a mini PC then the GMKtec EVO-X2 AI Mini PC AMD Ryzen™ AI Max+ 395 with 64GB RAM is advertised at $2K. But you’

Man, these things got expensive. I picked up a new GMKTec Evo-X2 for like $2,000 when it first came out. That made it an easy decision over the DGX Spark that was expected to be double the price. It certainly wasn't great early on, but software support is much better now. I did want to use it for ga

I have the 96GB version and i must say im happy with it. I can try many different models and get decent outputs. Also i think with MTP and other technology the futur will let you run good models on smaller hardware. I would still buy the 96GB version but i mean that the models on it could be better.

Te sale más barato comprar uno de esos pc con ryzen 395 Ai max con 128gb ram. Puedes destinar como 96gb para que corran los llm. Busca los gmktec Evo x2 Un Mac mini si pillas sería buena opción tb

I have this chipset in a GMKTec Evo X2 and tbh it doesn't need overclocking at all, BUT you can use Universaltuner to overclock or underclock, or just undervolt. It also allows you to do some serious tweaking like adjusting how fast it ramps up performance, or the boost calculations so that average

I have 3, 3 different machines. Same processor, same memory, 128Gb. https://www.gmktec.com/products/amd-ryzen%E2%84%A2-ai-max-395-evo-x2-ai-mini-pc?srsltid=AfmBOora2-dt5AKlG0sAXmT6LYm6bZ8GVZtxE5mRlfFN5SJn6JCKAQF9&variant=64bbb08e-da87-4bed-949b-1652cd311770 https://www.bee-link.com/products/beelink-

I run a AMD RYZEN AI MAX+ 395 w/ Radeon™ 8060S × 32 with 128GB of shared RAM and run a couple of models locally with llama-swap and OpenWebUI on Ubuntu 24.04 LTS. GPT-OSS 120b, Qwen3.5-122B or Qwen3-Coder-Next-80B all run reasonably fast. My GMKtec NucBox_EVO-X2 cost less than $2,000 last year in Au