Turing ULRv5

Training compute

2.9×10²² FLOP

Parameters

2.2B

Published

Sep 28, 2021

Turing ULRv5 is an AI model developed by Microsoft (United States), first published in September 2021. It works in the language domain, on tasks such as language modeling/generation, question answering and translation.

Training it took an estimated 2.9×10²² FLOP of compute (estimation method: hardware). The model has 2,200,000,000 parameters. Training ran on 256 NVIDIA A100 for about 336 hours.

Access: Hosted access (no API). Its weights are not openly released. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Microsoft
Country of organization: United States
Domain: Language
Task: Language modeling/generation, Question answering, Translation
Training compute: 2.9×10²² FLOP
Compute estimation method: Hardware
Parameters: 2,200,000,000
Training hardware: NVIDIA A100
Chips used: 256
Training time: 336 h
Training power draw: 206.7 kW
Model accessibility: Hosted access (no API)
Open weights: No
Epoch confidence: Confident