BigRNA

Training compute

1.2×10²² FLOP

Parameters

2B

Published

Sep 27, 2023

BigRNA is an AI model developed by DeepGenomics (Canada), first published in September 2023. It works in the biology domain, on tasks such as drug discovery, protein-rna binding affinity prediction and gene expression profile generation.

Training it took an estimated 1.2×10²² FLOP of compute (estimation method: operation counting). The model has 2,000,000,000 parameters. It was trained on roughly 1T datapoints.

Access: Unreleased. Its weights are not openly released. Epoch AI rates the confidence of this record as likely.

Full record

Organization: DeepGenomics
Country of organization: Canada
Domain: Biology
Task: Drug discovery, Protein-RNA binding affinity prediction, Gene expression profile generation
Training compute: 1.2×10²² FLOP
Compute estimation method: Operation counting
Parameters: 2,000,000,000
Dataset size: 1T
Model accessibility: Unreleased
Open weights: No
Epoch confidence: Likely