Evo 2

Models genomic sequences and can be used to score sequence variation and predict the effects of genetic variants.

Field
Biology & genomics
Method
Foundation model
Data it takes
sequence
License
Apache-2.0

Evo 2 is a genome-scale language model trained across the tree of life. It assigns a likelihood to a DNA sequence, which supports variant effect prediction: a variant that lowers sequence likelihood is a candidate for functional disruption. It also generates sequences and produces embeddings.

It runs on H100-class GPUs. Models at this scale are primarily an infrastructure problem rather than a scientific one.

Upstream project →

Catalog entry last checked 2026-08-17. All models →

Run Evo 2 on your data

The lab runs this on its own compute, in a container, with the inputs and parameters recorded alongside the result. Initial scoping conversations are free.

Contact the Lab