V-JEPA

Training compute

1.6×10²¹ FLOP

Parameters

630M

Published

Feb 15, 2024

V-JEPA is an AI model developed by Meta AI (United States), first published in February 2024. It works in the vision domain, on tasks such as representation learning.

Training it took an estimated 1.6×10²¹ FLOP of compute. The model has 630,000,000 parameters. Training ran on NVIDIA A100 for about 50 hours.

Epoch AI rates the confidence of this record as confident.

Full record

Organization: Meta AI
Country of organization: United States
Domain: Vision
Task: Representation learning
Training compute: 1.6×10²¹ FLOP
Parameters: 630,000,000
Training hardware: NVIDIA A100
Training time: 50 h
Epoch confidence: Confident