DataRater test model (1B)

Training compute

1.2×10²⁰ FLOP

Parameters

1B

Published

May 23, 2025

DataRater test model (1B) is an AI model developed by Google DeepMind (United States), first published in May 2025. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 1.2×10²⁰ FLOP of compute. The model has 1,000,000,000 parameters. Training ran on 32 Google TPU v6e Trillium.

The reference paper has 12 citations. Epoch AI rates the confidence of this record as likely.

Full record

Organization: Google DeepMind
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 1.2×10²⁰ FLOP
Parameters: 1,000,000,000
Training hardware: Google TPU v6e Trillium
Chips used: 32
Training power draw: 23.8 kW
Citations: 12
Epoch confidence: Likely