Ryzen AI Max 390

The Ryzen AI Max 390 is a 12-core / 24-thread processor based on the Strix Halo (2025) architecture, manufactured on a 4 nm process node. It reaches a maximum boost clock of 5 GHz from a base frequency of 3.2 GHz— higher boost clocks directly translate to better single-threaded performance and responsiveness in gaming. It carries 64 MB (total) of L3 cache, plus 1 MB (per core) of L2 cache and 80 kB (per core) of L1 cache per core. In the PassMark benchmark (a comprehensive multi-threaded test), it scores 41,834, placing it in the Enthusiast performance tier as a Current Gen generation product.

The Ryzen AI Max 390 fits into the FP11 socket and supports PCIe 4.0 with 28 total PCIe lanes for expansion cards and storage. It supports DDR5 memory at speeds up to 8000 (faster RAM improves bandwidth-sensitive workloads), with a maximum capacity of 128 across 4 channels (dual-channel doubles bandwidth). It supports ECC memory, providing error-correcting capabilities — essential for mission-critical workloads and servers. Integrated graphics: AMD Radeon 8050S — allows display output without a dedicated GPU, useful for troubleshooting and light tasks.

It features an unlocked multiplier, allowing enthusiasts to overclock beyond stock frequencies for additional performance. It supports AVX-512 instructions — specialized SIMD operations that significantly accelerate AI inference, scientific computing, and video encoding workloads. Virtualization: VT-x, VT-d, AMD-V — enables running virtual machines efficiently (important for development and server workloads). Its primary market competitor is the Apple M4 Max.

The Ryzen AI Max 390 has a rated TDP (Thermal Design Power) of 55 Watt — this indicates the amount of heat generated under sustained load and determines what cooling solution is needed. No stock cooler is included — you will need to purchase an aftermarket cooler separately. Recommended cooling: High-end Laptop Cooling.

Performance ranking position: #271 out of all indexed processors.