Contents

AMD ROCm™ Blogs

Ecosystems and partners

Stone Ridge Expands Reservoir Simulation Options with AMD Instinct™ Accelerators

Stone Ridge Technology latest development effort was to port ECHELON from CUDA to the AMD HIP platform, enabling ECHELON to use AMD Instinct GPUs like the MI210, MI250X, and the upcoming MI300 Series

Stone Ridge Expands Reservoir Simulation Options with AMD Instinct™ Accelerators
University of Michigan, AMD collaboration

AMD Collaboration with the University of Michigan offers High Performance Open-Source Solutions to the Bioinformatics Community

AMD Collaboration with the University of Michigan offers High Performance Open-Source Solutions to the Bioinformatics Community
Siemens and AMD partnership

Siemens taps AMD Instinct™ GPUs to expand high-performance hardware options for Simcenter STAR-CCM+

Siemens taps AMD Instinct™ GPUs to expand high-performance hardware options for Simcenter STAR-CCM+

Applications and models

Segment Anything with AMD GPUs

The Segment Anything Model (SAM) is a cutting-edge image segmentation model that democratizes promptable segmentation

./artificial-intelligence/segment-anything/README.html
Detectron2 on AMD GPUs

Panoptic segmentation and instance segmentation with Detectron2 on AMD GPUs

Panoptic segmentation and instance segmentation with Detectron2 on AMD GPUs
Accelerating Large Language Models with Flash Attention on AMD GPUs

In this blog post, we will guide you through the process of installing Flash Attention on AMD GPUs

Accelerating Large Language Models with Flash Attention on AMD GPUs
Step-by-Step Guide to Use OpenLLM on AMD GPUs

OpenLLM is an open-source platform designed to facilitate the deployment and utilization of large language models (LLMs)

Step-by-Step Guide to Use OpenLLM on AMD GPUs
Inferencing with Mixtral 8x22B on AMD GPUs

Mixtral 8x22B is a sparse MoE decoder-only transformer model, get it working on AMD GPUs

Inferencing with Mixtral 8x22B on AMD GPUs
Training a Neural Collaborative Filtering (NCF) Recommender

Collaborative Filtering is a type of item recommendation where new items are recommended to the user based on their past interactions.

Training a Neural Collaborative Filtering (NCF) Recommender on an AMD GPU

Software tools & optimizations

SmoothQuant model inference on AMD Instinct MI300X using Composable Kernel

The AMD ROCm™ Composable Kernel (CK) library provides a programming model for writing performance-critical kernels for machine learning workloads.

./software-tools-optimization/ck-int8-gemm-sq/README.html
AMD in Action: Unveiling the Power of Application Tracing and Profiling

Rocprof is a robust tool designed to analyze and optimize the performance of HIP programs on AMD ROCm platforms

AMD in Action: Unveiling the Power of Application Tracing and Profiling
Reading AMD GPU ISA

In this blog post, we will discuss how to read and understand the ISA for AMD’s Graphics Core Next (AMDGCN) architecture

Reading AMD GPU ISA
Application portability with HIP

HIP enables these High-Performance Computing (HPC) facilities to transition their CUDA codes to run and take advantage of the latest AMD GPUs

Application portability with HIP
C++17 parallel algorithms and HIPSTDPAR

The C++17 standard added the concept of parallel algorithms to the pre-existing C++ Standard Library

C++17 parallel algorithms and HIPSTDPAR
Affinity part 1 - Affinity, placement, and order

Affinity is a way for processes to indicate preference of hardware components so that a given process is always scheduled to the same set of compute cores and is able to access data from local memory efficiently

Affinity part 1 - Affinity, placement, and order

Stay informed