Projected GAN

Training compute

1.1×10¹⁹ FLOP

Published

Nov 1, 2021

Projected GAN is an AI model developed by Heidelberg University (Germany), first published in November 2021. It works in the image generation domain, on tasks such as image generation.

Training it took an estimated 1.1×10¹⁹ FLOP of compute (estimation method: hardware). It was trained on roughly 3M datapoints. Training ran on NVIDIA V100,NVIDIA Quadro RTX 6000.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 280 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Heidelberg University
Country of organization: Germany
Domain: Image generation
Task: Image generation
Training compute: 1.1×10¹⁹ FLOP
Compute estimation method: Hardware
Dataset size: 3M
Training hardware: NVIDIA V100, NVIDIA Quadro RTX 6000
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 280
Epoch confidence: Confident