XY-LENTXL

Training compute

7.2×10²¹ FLOP

Parameters

2B

Published

Oct 26, 2022

XY-LENTXL is an AI model developed by Microsoft (United States), first published in October 2022. It works in the language domain, on tasks such as representation learning and language modeling.

Training it took an estimated 7.2×10²¹ FLOP of compute (estimation method: operation counting). The model has 2,000,000,000 parameters. It was trained on roughly 600B datapoints. Training ran on 512 NVIDIA H100 SXM5 80GB.

Access: Unreleased. 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: Representation learning, Language modeling
Training compute: 7.2×10²¹ FLOP
Compute estimation method: Operation counting
Parameters: 2,000,000,000
Dataset size: 600B
Training hardware: NVIDIA H100 SXM5 80GB
Chips used: 512
Training power draw: 717.0 kW
Model accessibility: Unreleased
Open weights: No
Epoch confidence: Confident