AMD Investor Day Presentation Deck

CPU Stack WinML** ONNX Runtime Optimized Inference Models Pytorch Windows Runtime AMD AI SOFTWARE TODAY ML Graph Compiler ZenDNN, AOCL Optimized Library ZEN Studio (AOCC) Ryzen™ CPUs *Ryzen uses Radeon ML and WinML **WinML for Ryzen Only TensorFlow Linux Runtime EPYC™ CPUs ROCM™ Platform TM Optimized Inference Models ONNX Runtime Pytorch AMD Instinct™ GPUS MIGraphX: Al Development Tools MIGraphX: ML Graph Compiler MIOpen, ROCBLAS ROCM™ HIP Compiler and tools TensorFlow Runtime Radeon™ GPUS Al Development Tools: Quantizer, Sparsity/Pruning Tool Al Deployment Tools: Inference Server Vitis™ Al Platform Optimized Inference Models ONNX Runtime Pytorch Vitis Al Development & Deployment Tools Vitis Al ML Graph Compiler Vitis ML Libraries Vitis™ SW Platform (AIE Compiler, HLS and tools) CNN Overlay TensorFlow VersalⓇ Adaptive SoCs Runtime Transformer Overlay ZynqⓇ Adaptive SoCs

View entire presentation