Distilled Grandmaster

Training compute

10²² FLOP

Parameters

270M

Published

Feb 7, 2024

Distilled Grandmaster is an AI model developed by DeepMind (United Kingdom), first published in February 2024. It works in the games domain, on tasks such as chess.

Training it took an estimated 10²² FLOP of compute (estimation method: operation counting). The model has 270,000,000 parameters. It was trained on roughly 1.2T datapoints.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 44 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: DeepMind
Country of organization: United Kingdom
Domain: Games
Task: Chess
Training compute: 10²² FLOP
Compute estimation method: Operation counting
Parameters: 270,000,000
Dataset size: 1.2T
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 44
Epoch confidence: Confident