IEEE 3378-2025

This standard applies to the design, development, application, and evaluation of large-scale deep learning models for artificial intelligence. The standard specifies a technical reference architecture of large-scale deep learning models. Furthermore, the standard provides general requirements for basic software and hardware functions of these learning models and their performance. The standard also includes a framework for the evaluation of such large-scale deep learning models.



New IEEE Standard - Active. The tasks that large-scale deep learning models can support are specified in this standard. The evaluation metrics and evaluation requirements of large-scale deep learning models are addressed in this standard.