GPUDatacenterReleased —

AMD Instinct MI430X

The Instinct MI430X is AMD's HPC-focused accelerator in the Instinct MI400 series, built on the CDNA 5 architecture. It keeps full hardware FP64 throughput for scientific simulation alongside low-precision AI formats, with 432 GB of HBM4 memory. First announced in November 2025, it is expected to be available in 2027 and has been selected for supercomputers including Discovery at Oak Ridge National Laboratory and Alice Recoque in France.

General

Manufacturer: AMD
Model: Instinct MI430X
Segment: Datacenter
GPU chip: —
Architecture: CDNA 5
Release date: —
Price: Search on Amazon

Memory

Memory size: 432 GB
Memory type: HBM4
Bus width: —
Bandwidth: 23,300 GB/s
Memory clock: —

Clocks & silicon

GPU clock: —
Boost clock: —
Shader units: —
Tensor / AI cores: —
RT cores: —
Transistors: —
Technology (nm): 2 nm
TSMC N2
TDP / TBP: —

Interface & APIs

Bus interface: —
SLI / multi-GPU: —
DirectX: —
OpenGL: —
Vulkan: —
Other APIs: —

AI support

Precisions: FP8FP16BF16INT8INT4TF32FP64
AI accelerator: —
Peak AI TOPS: —
Notes: AMD rates the MI430X at up to 288 TFLOPS of hardware FP64 and up to 9.2 PFLOPS of FP4 for AI.

All AMD GPUs →

Datacenter gpus from AMD announced around the same time as the AMD Instinct MI430X. Add any of them to your compare list to see the specs side by side.

gpu