GNN

Training compute

1.6×10⁹ FLOP

Parameters

30

Published

Dec 9, 2008

GNN is an AI model developed by University of Siena (Italy), first published in December 2008. It works in the other domain, on tasks such as binary classification.

Training it took an estimated 1.6×10⁹ FLOP of compute (estimation method: operation counting). The model has 30 parameters. It was trained on roughly 207 datapoints.

Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as confident.

Full record

Organization: University of Siena
Country of organization: Italy
Domain: Other
Task: Binary classification
Training compute: 1.6×10⁹ FLOP
Compute estimation method: Operation counting
Parameters: 30
Dataset size: 207
Chips used: 1
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Epoch confidence: Confident