Open-source GPU compute software stack for AMD Instinct accelerators across HPC and AI workloads.
ROCm is AMD's open software platform for high-performance computing and AI on AMD Instinct GPUs, providing drivers, runtimes, compilers, and libraries that target accelerator workloads on Linux servers.
The stack includes HIP for portable GPU programming, math and communication libraries such as rocBLAS and RCCL, and integrations with major machine-learning frameworks, and competes with NVIDIA's proprietary CUDA platform in data center and research deployments.
| Attribute | ROCm | Aviator | CUDA, TensorRT | Intel oneAPI | |
|---|---|---|---|---|---|
| Provider |  |  |  |  | |
| Founded | 1969 | 2019 | 1993 | 1968 | |
| Sells To | Consumers, Enterprise | Enterprise | Consumers, Enterprise | Enterprise | |
| Pricing Model | Software, Transactional | — | Licensing, Product Sales, Subscription | Hardware Sales | |
| Ownership | Public | Private | Public | Public |
Provider
Founded 1969
Sells To Consumers, Enterprise
Pricing Model Software, Transactional
Ownership Public
Provider
Founded 2019
Sells To Enterprise
Pricing Model —
Ownership Private
Provider
Founded 1993
Sells To Consumers, Enterprise
Pricing Model Licensing, Product Sales, Subscription
Ownership Public
Provider
Founded 1968
Sells To Enterprise
Pricing Model Hardware Sales
Ownership Public
ONNX Runtime is a cross-platform inference and training accelerator for machine-learning models.
Low-level graphics and compute API for Apple silicon GPUs across iOS, iPadOS, macOS, and visionOS.
Software suite for developing and deploying physical AI applications on SiMa.ai hardware.
An open-source LLM inference engine with hand-tuned kernels and speculative decoding for consumer hardware.
Amazon Bedrock is AWS's managed service for building applications with foundation models.
Gemini Enterprise Agent Platform supports building, deploying, and governing enterprise AI agents and machine learning models.
Open-source platform for collaborating on machine learning models, datasets, and applications.
Tenstorrent liquid-cooled AI workstation for desk-side training and inference.
Tenstorrent scale-out server family for production AI training and inference.
Tenstorrent PCIe accelerator card family for AI training and inference workloads.
ONNX Runtime is a cross-platform inference and training accelerator for machine-learning models.
Low-level graphics and compute API for Apple silicon GPUs across iOS, iPadOS, macOS, and visionOS.
Software suite for developing and deploying physical AI applications on SiMa.ai hardware.
An open-source LLM inference engine with hand-tuned kernels and speculative decoding for consumer hardware.
Amazon Bedrock is AWS's managed service for building applications with foundation models.
Gemini Enterprise Agent Platform supports building, deploying, and governing enterprise AI agents and machine learning models.
Open-source platform for collaborating on machine learning models, datasets, and applications.
Tenstorrent liquid-cooled AI workstation for desk-side training and inference.
Tenstorrent scale-out server family for production AI training and inference.
Tenstorrent PCIe accelerator card family for AI training and inference workloads.