PaliGemma

Training compute

1.1×10²² FLOP

Parameters

3B

Published

Jul 10, 2024

PaliGemma is an AI model developed by Google DeepMind (United States), first published in July 2024. It works in the vision domain, on tasks such as visual question answering.

Training it took an estimated 1.1×10²² FLOP of compute (estimation method: hardware). The model has 3,000,000,000 parameters. Training ran on 256 Google TPU v5e,Google TPU v3 for about 162 hours.

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

Full record

Organization: Google DeepMind
Country of organization: United States
Domain: Vision
Task: Visual question answering
Training compute: 1.1×10²² FLOP
Compute estimation method: Hardware
Parameters: 3,000,000,000
Training hardware: Google TPU v5e, Google TPU v3
Chips used: 256
Training time: 162 h
Model accessibility: Open weights (restricted use)
Open weights: Yes
Epoch confidence: Confident