Image 1 of 7
Image 2 of 7
Image 3 of 7
Image 4 of 7
Image 5 of 7
Image 6 of 7
Image 7 of 7
Personal AI Appliance & Local Server Overview
Technical Specs
Private compute resources easily accessible from the web. A quiet, 24/7-ready workstation that brings private, local-first AI to your desk.
Larger-than-life performance, powered by Ryzen™ AI Max APU enabling a CPU + GPU with shared memory.
A seriously impressive CPU. Multi-task to the highest order with up to 16 cores and 5.1GHz max boost. Run legit, state-of-the-art AI models like Llama 70B right on your desk with up to 96GB of graphics addressable memory and a 256-bit memory bus.
Smaller than a Sheet of Paper: With a side area of 205 x 232 millimeters, the chassis is 0.76 times the size of a sheet of paper and 90 millimeters wide for a slim, sleek look.
The CI Core Server has 5Gbit Ethernet along with two USB4 ports, allowing networking multiple together to run even larger models with llama.cpp RPC
APU CPU:
AMD AI Max+ 395 - 128GB
3.0GHz base clock - Up to 5.1GHz max boost
16-core/32-thread 64MB L3 Cache
iGPU:
Radeon™ 8060S Graphics
40 Graphics Cores, up to 2.9GHz
Simultaneously supports up to 4 displays
Memory: 128GB LPDDR5x-8000
Storage: 2TB NVMe, Crucial
Power: ITX-PFC 500W
Networking:
Wi-Fi 7 - AMD RZ717
5Gbit Ethernet - Realtek RTL8126
Size:
4.3L
205× 232× 90mm
Front plate: 1.5mm thick steel. Rear plate: 1.2mm thick steel. Other panels: 0.8mm thick steel.
Front Ports:
1 USB 3.2
Rear Ports:
1x HDMI - HDMI 2.1
2x DisplayPort - DP 2.1 up to 10 Gb/s
2x USB-C - USB4
1x RJ45 - 5Gbit Ethernet
2x USB-A - USB 3.2 Gen 2
1x 3.5mm Combo
Connectivity:
Front panel audio
1x PCIe x4 slot
Framework Motherbard
Cooling:
Noctua NF-A12x25 Premium Quiet Fan 120mm, Brown
Heatsink 123.75 x 123.15 x 54.58mm 6x copper heatpipes Aluminum fins Honeywell PTM7958 thermal interface
Pictured Companion Can for scale, not included.