Weight Decay is an AI model, first published in December 1991. It works in the speech domain, on tasks such as speech synthesis. It counts among the frontier models: the systems trained with the most compute of their moment.
Training it took an estimated 7.5×10¹⁰ FLOP of compute (estimation method: operation counting). The model has 8,386 parameters. It was trained on roughly 25K datapoints.
Epoch AI rates the confidence of this record as confident.