Discover our Deep Learning Solutions

Radeon Instinct™ MI Series is the fusion of human instinct and machine intelligence, designed to be open from the metal forward. Higher levels of datacenter performance and efficiencies are enabled through AMD’s introduction of world-class GPU technologies and the Radeon Instinct’s open ecosystem approach to datacenter design through our ROCm software platform, support of various system architectures, and industry standard interconnect technologies.

Get started on one of our featured deep learning solutions now.

Radeon Instinct™ Server Accelerators for Deep Learning and HPC Applications

Evolving. Adapting. Learning.

Radeon Instinct™ Server Accelerators for Deep Learning and HPC Applications

Cloud/Hyperscale, Financial Services, Energy, Life Sciences, Other HPC Workloads

Radeon Instinct™ MI25 Accelerator

Radeon Instinct™ MI25 Accelerator

Superior FP16 and FP32 performance for machine intelligence and deep learning.
Compute Units:
64
Stream Processors:
4096
Peak Half Precision (FP16) Performance:
24.6 TFLOPs
Peak Single Precision (FP32) Performance:
12.29 TFLOPs
Peak Double Precision (FP64) Performance:
768 GFLOPs
Memory Size:
16 GB
Memory Type (GPU):
HBM2
Memory Bandwidth:
484 GB/s
Radeon Instinct™ MI8 Accelerator

Radeon Instinct™ MI8 Accelerator

Scalable accelerator for machine and deep learning inference applications.
Stream Processors:
4096
Peak Half Precision (FP16) Performance:
8.19 TFLOPs
Peak Single Precision (FP32) Performance:
8.19 TFLOPs
Peak Double Precision (FP64) Performance:
512 GFLOPs
Memory Size:
4 GB
Memory Type (GPU):
High Bandwidth Memory (HBM)
Memory Bandwidth:
512 GB/s
Radeon Instinct™ MI6 Accelerator

Radeon Instinct™ MI6 Accelerator

'Versatile accelerator for machine and deep learning training and inference applications.
Stream Processors:
2304
Peak Half Precision (FP16) Performance:
5.73 TFLOPs
Peak Single Precision (FP32) Performance:
5.73 TFLOPs
Peak Double Precision (FP64) Performance:
358 GFLOPs
Memory Size:
16 GB
Memory Type (GPU):
GDDR5
Memory Bandwidth:
224 GB/s
GPUOpen logo

Introducing GPUOpen

GPUOpen is an AMD initiative designed to enable developers to create ground-breaking computer generated imagery and GPU computing applications for great performance and lifelike experiences using no cost and open development tools and software.

ROCm logo

ROCm - A New Era in Open GPU Computing

Platform for GPU-enabled HPC and ultrascale computing.

AMD Deep Learning Solutions

Deep Learning Solutions

Accelerate discovery with optimized server solutions.

Epyc

A New Era in the Datacenter

AMD EPYC™ - technology for the software-defined datacenter.