MoLFormer-XL

Training compute

4.5×10²⁰ FLOP

Published

Jan 25, 2023

MoLFormer-XL is an AI model developed by IBM (United States), first published in January 2023. It works in the biology domain, on tasks such as protein generation and protein folding prediction.

Training it took an estimated 4.5×10²⁰ FLOP of compute (estimation method: hardware). Training ran on 16 NVIDIA V100 for about 208 hours.

Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as confident.

Full record

Organization: IBM
Country of organization: United States
Domain: Biology
Task: Protein generation, Protein folding prediction
Training compute: 4.5×10²⁰ FLOP
Compute estimation method: Hardware
Training hardware: NVIDIA V100
Chips used: 16
Training time: 208 h
Training power draw: 9.6 kW
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Epoch confidence: Confident