AMD Ryzen AI Max PRO 380 socket: FP11

The AMD Ryzen AI Max PRO 380 uses FP11, soldered with 2,077 balls. It is the 6-core PRO version of the Strix Halo APU with a 55 W default TDP, and the soldered chip cannot be replaced or upgraded.

Socket
FP11
Launched
January 2025
Cores and threads
6 and 12
Default TDP
55 W
Codename
Strix Halo
Segment
Mobile (soldered)
Clocks
3.6 / 4.9 GHz

AMD Ryzen AI Max PRO 380 specifications

Cores6 (uniform Zen 5 cores)
Threads12
Core architectureZen 5, chiplet CCDs with large I/O die
Clocks3.6 GHz base, up to 4.9 GHz maximum boost
Cache16 MB L3 and 6 MB L2 (22 MB total)
Default TDP55 W
Configurable TDP45 to 120 W
Manufacturing processTSMC 4 nm
Memory supportSoldered LPDDR5X up to 8000 MT/s, 256-bit bus, up to 128 GB unified
PCI Express16 PCIe Gen4 lanes
Integrated graphicsAMD Radeon 8040S Graphics, 16 compute units, up to 2800 MHz
NPUXDNA 2, up to 50 TOPS
PackageFP11 (BGA2077, soldered)
Cooler in the boxNone (soldered, system thermal solution required)

Motherboard compatibility

AMD Ryzen AI Max PRO 380 needs a motherboard with a FP11 socket. Chipset by chipset data for this codename is not recorded yet. See the FP11 chipset list and the board maker's CPU support list.

FP11 at a glance

TypeBGA, 2,077 contacts
Package size37.5 × 45 mm
IntroducedJanuary 2025
MemorySoldered LPDDR5X only, 256-bit bus (8 channels), up to LPDDR5X-8533 and up to 128 GB of unified memory shared by CPU, GPU and NPU, with Link-ECC on embedded boards
Cooler mounting holesNot applicable (soldered)

Read the full FP11 socket guide

What the Ryzen AI Max PRO 380 is

The Ryzen AI Max PRO 380 is a 6-core, 12-thread mobile workstation APU that AMD launched in January 2025 as the smallest PRO SKU of the Strix Halo silicon. Its six uniform Zen 5 cores run at 3.6 GHz and boost to 4.9 GHz, backed by 6 MB of L2 and 16 MB of L3 cache. Graphics come from AMD Radeon 8040S Graphics with 16 compute units at up to 2800 MHz, and an XDNA 2 NPU delivers up to 50 TOPS for local AI work.

The PRO suffix marks the business stack. The chip carries AMD PRO Technologies, the manageability and security feature set that IT departments use to deploy and maintain fleets of laptops and mobile workstations. The silicon class matches the consumer Halo parts, so a company laptop gets the same large integrated GPU and unified memory pool as the retail designs, with the PRO layers added for fleet use.

Strix Halo is a chiplet design. Zen 5 core chiplets sit beside a large I/O die that holds the graphics engine, the NPU and the memory controller. AMD builds the compute silicon on the TSMC 4 nm process. The laptop maker sets the cooling and power policy inside a 45 to 120 W configurable window, with 55 W as the default TDP for this SKU.

With six CPU cores, the 380 suits buyers who need the Halo memory subsystem more than thread count. The 256-bit LPDDR5X interface feeds up to 128 GB of unified memory that the CPU, GPU and NPU share, so large models and data sets stay resident without copies between separate memories. It is the smallest core count in the Halo PRO range for designs that prize graphics memory and AI throughput over multicore rendering.

Socket and motherboard compatibility

The Ryzen AI Max PRO 380 is soldered to its board in the FP11 ball grid array, so there is no socket to plug into and the chip cannot be replaced or upgraded later. The 2,077-ball joint is made once at the factory on a 37.5 × 45 mm package, and the laptop design is qualified around this exact SKU for its whole service life.

There is no board or chipset choice in the retail sense. The laptop vendor selects the memory capacity, tunes the firmware and validates the thermal design for the 45 to 120 W window. Moving from the 380 to the 8-core AMD Ryzen AI Max PRO 385 or the 12-core AMD Ryzen AI Max PRO 390 means a different board build, not a part swap, even though all three share the FP11 package.

The FP11 package is shared with the embedded branch of the same silicon. The industrial AMD Ryzen AI Embedded X188i uses the same 2,077-ball footprint for equipment with long lifecycles, while the smaller FP8 package serves lower power Strix Point designs. That split keeps Halo boards distinct from mainstream thin laptop boards.

Memory, PCIe and cooling

Memory is soldered LPDDR5X up to 8000 MT/s on a 256-bit bus, with up to 128 GB in one unified pool shared by the CPU cores, the 16 compute unit graphics engine and the NPU. Unified memory removes the copy step between a separate graphics card and system RAM, which helps large AI models and GPU compute tasks that exceed the size of a typical laptop graphics allocation. Capacity and speed are fixed when the board is built, so they rank with the SKU as permanent choices.

Expansion comes from 16 PCIe Gen4 lanes from the SoC, which typically serve NVMe storage and high speed peripherals, plus USB4 ports for docks and external devices. The integrated Radeon graphics handle display and parallel compute, so the lanes stay free for storage and I/O. The exact lane split and port list are set per laptop design, so the vendor specification governs.

Cooling is a system responsibility across the 45 to 120 W configurable range. The vendor sizes the heatsink and fan policy for the chassis, and there is no boxed cooler and no retail heatsink list for a soldered mobile APU. Slim workstation designs tend to run this SKU near the lower end of the window, while larger chassis allow higher sustained clocks.

Upgrade paths

There is no CPU upgrade path on a soldered package. When the workload outgrows six cores, the step inside FP11 is a different board build with the 8-core AMD Ryzen AI Max PRO 385 or the 12-core AMD Ryzen AI Max PRO 390, keeping the same LPDDR5X memory design with the power envelope scaled to the chassis.

Moving off FP11 means a new platform. The FP8 family serves lower power laptops with smaller integrated graphics, and desktop builders meet related Zen 5 silicon on socketed boards. Because memory is soldered on FP11, even a memory size change means a new board, so capacity planning belongs at the start of the project.

Frequently asked questions

What socket does the Ryzen AI Max PRO 380 use?

It uses FP11, a 2,077-ball grid array. The chip is soldered and cannot be moved to another board.

What is the difference between the PRO 380 and the consumer Halo chips?

The PRO 380 adds AMD PRO Technologies for business fleets, with manageability and security features for IT departments. The CPU core count, graphics configuration and memory subsystem follow the same Strix Halo design as the retail SKUs.

What memory does the PRO 380 use?

Soldered LPDDR5X up to 8000 MT/s on a 256-bit bus, up to 128 GB shared by the CPU, GPU and NPU. The size is fixed at board manufacture.

Does the PRO 380 include an NPU?

Yes, an XDNA 2 NPU with up to 50 TOPS, alongside 16 RDNA 3.5 graphics compute units and the six Zen 5 CPU cores.

Sources and verification

  1. AMD Ryzen AI Max PRO 380 (AMD)
  2. AMD announces expanded consumer and commercial AI PC portfolio at CES 2025 (AMD newsroom)
  3. AMD Ryzen AI Max PRO 380 benchmarks and specs (Notebookcheck)

Facts last checked against these sources on . Page last edited on .