End-to-end LLMs on Intel® Gaudi® 2 AI Accelerators: fine-tune 70B parameter LLaMA with LoRA, run real-time inference on 7B parameter LLaMA.

MD simulation with AI-based model for 1 million gold atoms using future Intel® Xeon® Processors codenamed Granite Rapids with MCR memory.

HPC/AI workloads running simultaneously on the same system with Intel® Data Center GPU Max 1100 using oneAPI for performance & versatility.

Circulating tumor cells computed by HARVEY visualized on Aurora with Intel® Data Center Max GPUs and Intel® oneAPI Rendering Toolkit.

Reconstruction of Protein-Protein Interaction networks by scaling Large Language Model (LLM) Inference on the Aurora Supercomputer.