Big Transfer (BiT-L)

Parameters

928M

Published

Dec 24, 2019

Big Transfer (BiT-L) is an AI model developed by Google Brain (United States), first published in December 2019. It works in the vision domain, on tasks such as image classification.

Epoch AI has no training-compute estimate for this model. The model has 928,000,000 parameters. It was trained on roughly 300M datapoints. Training ran on Google TPU v3.

Access: Unreleased. Its weights are not openly released. The reference paper has 1,361 citations.

Full record

Organization: Google Brain
Country of organization: United States
Domain: Vision
Task: Image classification
Parameters: 928,000,000
Dataset size: 300M
Training hardware: Google TPU v3
Numerical format: FP32
Model accessibility: Unreleased
Open weights: No
Citations: 1,361