Optimized GPU Software Stack
AMD ROCm™ is an open software platform that provides the drivers, libraries, compilers, developer tools, and APIs needed to build and deploy high-performance AI and HPC workloads on AMD GPUs. Optimized across the full software stack from low-level kernels to distributed applications. ROCm software delivers scalable performance while supporting familiar frameworks, open standards, and streamlined migration of existing GPU-accelerated code.
ROCm.AI: Accelerating Development Velocity and Performance Optimization
An AI-native developer experience that accelerates ROCm development, simplify AI workload development and optimization, and maximize performance on AMD hardware
AI Development: Build, Train, and Deploy at Scale with AMD ROCm
AMD ROCm™ software brings together open frameworks, optimized libraries, model tools, and scale-out infrastructure to support AI development from experimentation to production.
Simplified Model Development
Develop with PyTorch, TensorFlow, and JAX using AMD-optimized containers and workflows for popular open models, more than 3M+ from Hugging Face.
AMD Primus - Train AI Models at Scale
Use AMD Primus and ROCm communication libraries for pre-training, fine-tuning, and distributed training across AMD Instinct™ GPUs.
Optimize and Serve AI Models
Use AMD Quark, AITER, vLLM, and SGLang to optimize models, accelerate critical operations, and deploy high-throughput inference.
Deploy Across AI Infrastructure
Scale from individual GPUs to multi-node environments with containers, Kubernetes, Slurm, telemetry, and lifecycle-management tools.
HPC Developement: Build, Optimize, and Scale Scientific Computing with AMD ROCm
AMD ROCm™ software helps developers, researchers, and infrastructure teams turn AMD GPU capability into sustained application performance and faster time to scientific insight. Open programming models, optimized compute and communication libraries, and tools for profiling, validation, and multi-node deployment support workloads from individual kernels to leadership-class systems.
Accelerate Scientific Discovery
Accelerate scientific workloads with optimized libraries for linear algebra, FFTs, solvers, random-number generation, and parallel primitives. Profiling and debugging tools help identify bottlenecks and improve time to solution.
Open Development. Practical Portability.
Bring familiar HPC applications to AMD GPUs using HIP, OpenMP® offload, C/C++, Fortran, Python, and MPI. ROCm’s open-source foundation helps reduce migration friction and preserve flexibility.
Scale from GPU to Supercomputer
Scale applications across GPUs and nodes using ROCm-aware MPI, RCCL, rocSHMEM, and optimized data-transfer paths. Coordinate compute and communication while maintaining efficient data movement.
Proven at Leadership Scale
ROCm supports leadership-class systems including El Capitan, Frontier, and LUMI. These deployments demonstrate its scalability and programmability for demanding scientific computing environments.
What’s New
ROCm 10.0: Built for the Age of Agentic AI
Build with AMD Expertise
AMD Skills brings curated AMD knowledge and validated workflows into AI coding agents, helping developers build AMD-optimized applications with guidance in the tools they already use.
Run with Simpler Workflows
The ROCm CLI provides a unified interface to set up, manage and operate AI workloads on AMD hardware.
Optimize End-to-End Inference
ROCm Hyperloom uses agentic AI to profile workloads, identify bottlenecks, implement optimizations and validate performance across host code and GPU kernels.
Build on a Modular ROCm Core SDK
The ROCm Core SDK provides a modular foundation for building, deploying and optimizing AI workloads across AMD platforms.
Ecosystem and Partners
A broad, open AI ecosystem built with leading frameworks, technology partners, and developers to accelerate innovation, enable portability, and scale AI from development to production.
Supported Hardware
Developer Resources
ROCm Newsletter
Receive the latest ROCm news.
Footnotes
©2024 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD ROCm, AMD Instinct, EPYC, Radeon Instinct, and combinations thereof are trademarks of Advanced Micro Devices, Inc. PyTorch is a trademark or registered trademark of PyTorch. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.
- For a full list of Radeon parts supported by ROCm, go to https://rocm.docs.amd.com/en/latest/reference/gpu-arch-specs.html
Footnotes
©2024 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD ROCm, AMD Instinct, EPYC, Radeon Instinct, and combinations thereof are trademarks of Advanced Micro Devices, Inc. PyTorch is a trademark or registered trademark of PyTorch. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.
- For a full list of Radeon parts supported by ROCm, go to https://rocm.docs.amd.com/en/latest/reference/gpu-arch-specs.html