… This delivers measurable gains in kernel-level performance and runtime stability while reducing a model’s overall memory footprint. These updates improve out-of-the-box behavior in Windows ML inference workloads running on AMD GPUs. …
… For customers looking to run models with multiple batch sizes depending on the data input, this new feature now allows them to use a single compiled model to support multiple batch sizes, streamlining machine learning workflows, reducing storage space requirements, and reducing the script logic req… …
… Performance numbers are frequently updated as a reference point to help users test observed performance. See Performance Results Deployment and Orchestration Streamline and automate the deployment, monitoring, and maintenance of ML models in the data center. …
… July 23, 2026 Benchmarking AI Systems: from Model Metrics to Real-World Performance The agentic AI stack has evolved to fast multi-model orchestration, tool-augmented reasoning, and long-running inference chains. …
… What’s New in 2025.2 Enhanced Design Flow with AMD Versal™ AI Engines AI Engine API enhancements New and enhanced data types Programming model updates and optimizations New and enhanced DSP library functions Improved AI Engine mapper and router Easier Verification of Versal AI Engine Designs Update… …
… An aerospace and defense team can build on familiar x86 software and virtualization models. A broadcast, AV or casino gaming system builder can deploy new codecs, rendering updates and application workflows through software instead of waiting for a new fixed-function hardware cycle. …
… May 07, 2026 vLLM-ATOM: Unlocking Native AMD Performance in the vLLM Ecosystem — ROCm Blogs Use ATOM as an out-of-tree vLLM plugin to keep vLLM compatibility while enabling AMD-optimized attention, model execution, and multi-model support including Kimi-K2.5. …
… For Optimization “Optimize inference performance on this running model.” ROCm Hyperloom profiles the workload, identifies bottlenecks across host code and GPU kernels, applies targeted optimizations and validates performance and correctness. …
… This global competition challenges developers, researchers, and performance engineers to push the limits of large language model LLM inference performance on open models optimized for AMD Instinct™ MI355X GPUs. …