ByT5-XXL

Training compute

8.1×10²² FLOP

Parameters

12.9B

Published

May 28, 2021

ByT5-XXL is an AI model developed by Google and Google Research (United States), first published in May 2021. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 8.1×10²² FLOP of compute (estimation method: operation counting). The model has 12,900,000,000 parameters. It was trained on roughly 1T datapoints. Training ran on Google TPU v3. The compute alone is estimated at $92K in 2023 dollars.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 688 citations. Epoch AI rates the confidence of this record as likely.

Full record

Organization: Google, Google Research
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 8.1×10²² FLOP
Compute estimation method: Operation counting
Parameters: 12,900,000,000
Dataset size: 1T
Training hardware: Google TPU v3
Training cost (2023 USD): $92K
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 688
Epoch confidence: Likely