Newcomer Vzmore is building its flagship AX9 Max around the same motherboard platform found inside the Geekom A9 Max, but the German-owned brand is betting that a fundamentally different thermal design can extract better sustained performance from identical silicon. The AX9 Max pairs AMD's Ryzen AI 9 HX 470, a 12-core, 24-thread processor built on Zen 5 and Zen 5c cores, with integrated Radeon 890M graphics, 32 GB of DDR5 RAM, and a 1 TB SSD. All of it fits inside a chassis measuring 13.5 x 13.2 x 5.8 cm (5.3 x 5.2 x 2.3 inches) and weighing roughly 0.8 kg (1.7 lbs), with connectivity matching its Geekom counterpart: two USB4 ports and dual 2.5 Gbps Ethernet.

The marquee feature is Vzmore's proprietary V-Cooling system, which replaces conventional heatpipe layouts with a large vapor chamber paired with high-density cooling fins and a 360-degree bottom air intake. The company claims the design sustains a 65 W power limit while holding fan noise to 38 dB(A) under full load, a direct response to criticisms that the Geekom A9 Max runs loud when pushed hard. Early reviews from Guru3D and WCCFtech report CPU temperatures staying below 84°C (183°F) under sustained workloads, suggesting the cooling claims hold up in practice.

For Linux users, the underlying hardware is well supported on current distributions. The Radeon 890M's RDNA 3.5 architecture delivers strong OpenGL and Vulkan performance under the open-source Mesa and amdgpu kernel driver stack, with meaningful gains over the previous-generation 780M. The Ryzen AI 9 HX 470's NPU is also accessible on Linux through AMD's XDNA kernel driver and ROCm 7.2.1, opening the door to local AI inference workloads without a discrete GPU. Combined with dual 2.5 GbE and USB4, the AX9 Max could serve double duty as a compact home server or development box. That home-server angle has been tested directly on this unit: a review from Home & Smart installed the open-source Proxmox hypervisor on the AX9 Max and ran Ollama with the Qwen3.6-35B-A3B model inside a container, sustaining roughly 29 tokens per second by routing inference through the Radeon 890M, versus just 1 to 2 tokens per second when the test fell back to CPU-only inference.

The Vzmore AX9 Max ships with Windows 11 Pro and is available now starting at $1,500 (€1,380) for the 32 GB/1 TB configuration.