AMD enables AI agents with Ryzen platforms
AMD has introduced the Ryzen AI Halo developer platform alongside Ryzen AI Max PRO 400 Series processors to support next-generation agentic AI systems. Both platforms are designed to run AI workloads locally, reducing reliance on cloud infrastructure while supporting low-latency operation, real-time response, and large memory footprints. AMD says the platforms target AI-enabled PCs and workstation-class systems that can execute complex AI workflows, including autonomous agents, on a single machine.
The Ryzen AI Halo developer platform is powered by Ryzen AI Max+ 395 processors and provides up to 128 GB of unified system memory, allowing developers to run AI models with up to 200 billion parameters locally. Designed as a compact AI development system, it supports both Linux and Windows environments and works with frameworks including PyTorch, vLLM, llama.cpp, Ollama, ComfyUI, and LM Studio. AMD ROCm software optimization further supports local execution of large language models, diffusion models, and agent-based workflows.
Based on AMD’s Zen 5 architecture, Ryzen AI Max PRO 400 Series processors combine RDNA 3.5 graphics, an XDNA 2 NPU, up to 192 GB of system memory, and 160 GB of VRAM. The processors target commercial AI PCs, mobile workstations, and compact desktop systems handling AI, simulation, and visualization workloads.
Ryzen AI Halo systems will be available through Micro Center, with pre-orders starting in June 2026. AMD plans to expand the platform in the third quarter of 2026 with Ryzen AI Max PRO 400 Series processors.
The post AMD enables AI agents with Ryzen platforms appeared first on EDN.
