Fortress
A plug-and-play on-premises AI server. Its compact and modular design allows you to keep your data and AI workflows off the cloud and on your desk, ensuring complete privacy and control. It delivers enterprise-grade AI and data tooling to all, without the need for complex server infrastructure.
Fortress efficiently packs powerful AI compute to your desktop.
gpt-oss-20b (MKFP4) at 50+ tokens/sec
gpt-oss-120b (MKFP4) at 40+ tokens/sec
All at a fraction of the cost of an industry-grade AI server.
Specifications
AMD Ryzen AI Max+ 395 APU (16 cores / 32 threads, Zen 5 architecture)
Integrated Radeon 8060S GPU (40 RDNA 3.5 compute units)
XDNA 2 NPU with up to 126 TOPS for AI inference
128 GB LPDDR5x-8000 memory (soldered)
2 TB NVMe SSD pre-installed; supports dual M.2 PCIe 4.0 slots up to 8 TB total
Dual 10 GbE LAN ports
USB-C / USB4 ports (40 Gbps) and USB3.2 ports
WiFi 7 + Bluetooth 5.4 support
AI cluster
Scale, simply.
Upgrade your system by seamlessly connecting more nodes into a cluster.
More compute, more users, more possibilities.
