AMD will provide ROCm support for RDNA 3 consumer graphics cards, starting with Radeon RX 7900 XTX and Pro W7900
Thanks to CTOnews.com netizens Wu Yanzu in South China for the delivery of clues! CTOnews.com, July 2, ROCm is an open source software platform that allows researchers to leverage the potential of AMD Instinct accelerators to promote cross-platform high-performance computing and AI innovation.
As more and more developers start running ROCm on Radeon and Radeon PRO GPU to handle AI and ML work, AMD has released the latest ROCm 5.6 and announced that it will provide ROCm support for consumer graphics cards.
It is said that AMD will first adapt the ROCm for the two RDNA 3 GPU models, the RX 7900 XTX and the Pro W7900, and the official version will be launched this fall (ROCm 5.6). More models will be added later.
This is definitely big news for AMD users, especially those who need AI productivity. Not only that, CTOnews.com notes that ROCm 5.6 will bring a series of new features and enhancements, especially for AI and HPC workloads, including:
Integrate Hugging Face unit test suites into ROCm QA
Gradually support OpenAI Triton in PyTorch 2.0 inductor mode
Enable OpenXLA support for PyTorch, TensorFlow, and JAX through ROCm
ROCm 5.6 also introduces improvements to math libraries such as FFT, BLAS, and solvers, which are the basis of HPC applications, and enhancements to ROCm's development and deployment tools, including installation, ROCgdb (CPU-GPU integrated debugger), ROCm parser, and documentation.
AMD also said that it will continue its efforts to further optimize the framework and back-end compilers for better performance, including MLIR infrastructure improvements, as the basis for AMD's support for OpenAI Triton and OpenXLA compilers, and will continue to add more open source AI models optimized for AMD solutions to Hugging Face's AMD hub.
For HPC users, AMD has provided some solutions on AMD Infinity Hub so that customers can better build HPC application containers, and future versions of ROCm will increase the number of HPC applications supported on AMD Instinct solutions.
