NLLB

Training compute

1.8×10²² FLOP

Parameters

54.5B

Published

Jul 6, 2022

NLLB is an AI model developed by Meta AI (United States), first published in July 2022. It works in the language domain, on tasks such as translation.

Training it took an estimated 1.8×10²² FLOP of compute (estimation method: hardware). The model has 54,500,000,000 parameters. It was trained on roughly 300B datapoints. Training ran on NVIDIA A100 SXM4 80 GB. The compute alone is estimated at $51K in 2023 dollars.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 1,569 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Meta AI
Country of organization: United States
Domain: Language
Task: Translation
Training compute: 1.8×10²² FLOP
Compute estimation method: Hardware
Parameters: 54,500,000,000
Dataset size: 300B
Training hardware: NVIDIA A100 SXM4 80 GB
Chip-hours: 59.2K
Training cost (2023 USD): $51K
Numerical format: FP16
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 1,569
Epoch confidence: Confident