
Nucleotide Transformer
Foundation models for human genomics trained on DNA from thousands of genomes.
About
The Nucleotide Transformer is a collection of genomic foundation models from InstaDeep, NVIDIA, and the Technical University of Munich, introduced via bioRxiv in January 2023. The models are pre-trained on DNA sequences from over 3,200 diverse human genomes plus 850 genomes from other species.
Across 19 benchmarks the models matched or beat task-specific methods on 15, covering molecular phenotype prediction and other genomics tasks. Nine model sizes were open-sourced, from 500M parameters up. The release uses a CC BY-NC-SA 4.0 license, so commercial use requires permission.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 110; Trending score 76/100 from researcher assessment.