The Open-Source AI Stack
RSS
All hardware

Hardware · x86 unified

AMD Ryzen AI Max+ 395 (Strix Halo)

AMD

The first serious x86 unified-memory contender, up to 128 GB (about 96 GB convertible to VRAM) over a 256-bit LPDDR5X bus. Ships in mini-PCs and the Framework Desktop.

CPU · GPU · NPU 256 GB/s memory bus 128 GB Unified memory
x86 unified. Bus width tracks bandwidth (256 GB/s, sets decode speed); the box tracks capacity (128 GB, sets what fits). Unified memory is one pool shared by CPU and accelerator.

Specs

Memory
128 GB Unified LPDDR5X
Bandwidth
256 GB/s
NPU
50 INT8 TOPS
Power
120 W
Form factor
soc
Interconnect
none
Released
2025-01

What it runs (single unit, Q4_K_M, 4K context)

Model Fits? Decode ceiling
Llama 3.1 8B Instruct yes ~51 tok/s
Llama 3.3 70B Instruct yes ~6.3 tok/s
Qwen 2.5 72B Instruct yes ~6.1 tok/s
DeepSeek-V3 no —

Ceiling is the theoretical rooflineruntimeA performance model that bounds throughput by either compute or memory bandwidth, whichever is the limiting resource for an operation's arithmetic intensity. Open full entry ; open the explorer to set quant, context, and runtime and see the realistic range.

Sources