Pluribus

Training compute

6.6×10¹⁶ FLOP

Published

Jul 11, 2019

Pluribus is an AI model developed by Facebook AI Research (United States and France), first published in July 2019. It works in the games domain, on tasks such as poker.

Training it took an estimated 6.6×10¹⁶ FLOP of compute (estimation method: hardware).

Access: Unreleased. Its weights are not openly released. The reference paper has 797 citations. Epoch AI rates the confidence of this record as likely.

Full record

Organization: Facebook AI Research
Country of organization: United States, France
Domain: Games
Task: Poker
Training compute: 6.6×10¹⁶ FLOP
Compute estimation method: Hardware
Model accessibility: Unreleased
Open weights: No
Citations: 797
Epoch confidence: Likely