适用于AI工作负载的不可知软件将软件与硬件解耦

Agnostic Software for AI WorkloadsDecoupling Software from Hardware2Change is ComingAdapt or…3Workloads are Evolving at an Accelerated PaceModels are increasing in size exponentially, with no foreseeable slow downNetwork architectures are changing rapidlyTransformers are not the only solution that will existThe balance between training and inference is 50/50, even at scaleThe progression from ML to AI to Agents is not slowing downInsight:Observations:4Machine Learning is the Gateway to Autonomous AgentsSoftware 1.0Rule based and deterministic1957Software 2.0Data driven and discriminative2011Software 3.0Data driven and generative2018Software 4.0Goal-Directed agents2024Software 5.0Semi-Autonomous agents2030Software 6.0General Purpose agents20365Can Hardware Architecture Evolution Keep Up?Industry is hungry for a GPU alternative that has better efficiencyWe’ve reached the limit of physics for improved efficiencyEven at max capacity, Fabs cannot supply enough hardware, GPU or other