Accelerating AI with Open Software: AMD ROCm™ 7 is Here
…In the second half of 2025, AMD plans to enable developers to develop with ROCm locally on their own Linux or Windows device, with their own data, instead of relying on the…
Search
Test AI models and workflows with AMD Instinct GPUs on TensorWave
…Evaluate AMD Instinct GPUs on TensorWave Next-Gen AI Infrastructure Request Evaluation Proven Success Estimating LLM Inference Memory Requirements How to Run Multi-Node Training with Pyxis + AMD GPUs Sovereign AI: Infrastructure…
AMD AI Solutions
…May 10, 2026 For Developers Find the resources you need to develop your AI applications using AMD products. Access software downloads, documentation portals, training resources, communities, support forums, and more. AMD Developer…
Agentic AI Changes the CPU/GPU Equation
…AMD News Related Blogs View All Blogs Accelerating LLM Startup on AMD Ryzen AI with Two-Phase Custom Op Initialization In this blog, we describe a two-phase deferred initialization technique that…
Solutions for an AI-Ready Enterprise
…Modern Infrastructure for the AI Era Unlocking Business Value from Cloud to Client The AI Ready Enterprise 2 Plan and Build for Scale For organizations developing their AI strategy and deployment roadmap…
AMD AI DevDay 2026
…Member Experience AMD AI Developer Program members unlock the best DevDay experience, including early registration alerts, priority access where available, and member-only on-site perks. Not a member yet? Learn more…
Next Gen Networking Transport for Large Scale AI Training
…In addition, AMD developed the IB/RDMA transport semantic layer extensions for MRC, enabling seamless integration with existing RDMA programming models while introducing the multipath capabilities that distinguish MRC from traditional transports…
ZenDNN 5.2.1: Deepening Quantization and Expanding the AI Inference Frontier on AMD EPYC™ CPUs
…Asymmetric 4-bit WOQ allows more aggressive model compression while preserving accuracy, enabling production deployment of quantized LLMs on CPU infrastructure. Dynamic Quantization at Serving Time : The new INT8 dynamic quantization path…
Accelerating GPT-OSS-20B on AMD Ryzen™ AI NPUs: Efficient MoE Inference on Strix and Halo
…Enabling 20B LLMs on AMD Ryzen AI GPT-OSS-20B [1] is a 20B-parameter open-weight language model built for strong instruction following, coding, and general reasoning while remaining practical for…
AMD ROCm™ Software
AMD ROCm™ Software Open software stack that includes programming models, tools, compilers, libraries, and runtimes for AI and HPC solution development on AMD GPUs What’s New in ROCm 7 Access Latest…