DLRM-2022

Training compute

1.1×10²¹ FLOP

Parameters

3T

Published

Sep 15, 2021

DLRM-2022 is an AI model developed by Facebook (United States), first published in September 2021. It works in the recommendation domain, on tasks such as recommender system.

Training it took an estimated 1.1×10²¹ FLOP of compute (estimation method: reported). The model has 3,000,000,000,000 parameters. Training ran on NVIDIA Tesla V100 DGXS 32 GB,NVIDIA A100.

Access: Unreleased. Its weights are not openly released. The reference paper has 197 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Facebook
Country of organization: United States
Domain: Recommendation
Task: Recommender system
Training compute: 1.1×10²¹ FLOP
Compute estimation method: Reported
Parameters: 3,000,000,000,000
Training hardware: NVIDIA Tesla V100 DGXS 32 GB, NVIDIA A100
Model accessibility: Unreleased
Open weights: No
Citations: 197
Epoch confidence: Confident