AMD
GPUIntegrated (APU)Released Jan 6, 2025

AMD Radeon 8060S

256-bit · 2,560 cores

The Radeon 8060S is the full 40-compute-unit configuration of AMD's RDNA 3.5 Strix Halo integrated graphics. It shares a 256-bit LPDDR5X-8000 unified memory pool of up to 128 GB with the CPU, a large share of which can be assigned to the GPU, so the platform can run large AI models locally as well as demanding games and creative work that would otherwise need a discrete GPU. It powers the Ryzen AI Max+ 395 and Max+ PRO 395, and from 2026 the lower-core-count Ryzen AI Max+ 392 and 388.

General

Manufacturer
AMD
Model
Radeon 8060S
Segment
Integrated (APU)
GPU chip
Strix Halo
Architecture
RDNA 3.5
Release date
Jan 6, 2025

Memory

Memory size
—
Memory type
LPDDR5X (shared)
Bus width
256-bit
Bandwidth
256 GB/s
Memory clock
—

Clocks & silicon

GPU clock
—
Boost clock
2,900 MHz
Shader units
2,560 Stream Processors (40 CU)
Tensor / AI cores
—
RT cores
—
Transistors
—
Technology (nm)
4 nm
TSMC N4P
TDP / TBP
—

Interface & APIs

Bus interface
Integrated (on-die)
SLI / multi-GPU
—
DirectX
12 Ultimate (12_2)
OpenGL
4.6
Vulkan
1.3
Other APIs
OpenCL 2.1, HIP / ROCm, AV1 encode & decode

AI support

Precisions
FP8FP16BF16INT8INT4TF32FP64
AI accelerator
RDNA 3.5 WMMA matrix instructions + 40 Ray Accelerators
Peak AI TOPS
—
Notes
RDNA 3.5 supports WMMA matrix operations in FP16, BF16, INT8 and INT4 on the shader ALUs, without dedicated matrix cores. Up to 128 GB of unified LPDDR5X memory lets the GPU run large language models locally; the XDNA 2 NPU in the host APUs is a separate 50 TOPS block.
All AMD GPUs →

Integrated (APU) gpus from AMD announced around the same time as the AMD Radeon 8060S. Add any of them to your compare list to see the specs side by side.