Arc Institute · 3× · vu le 09 août 2026
Evo is a biological Foundation Model (genome language model) developed by Arc Institute, trained on DNA, RNA, and protein sequences, enabling both prediction and generative design tasks from individual molecules to entire genomes. The current version Evo 2 was developed in collaboration with NVIDIA, Stanford, UC Berkeley, and UCSF, features up to 40 billion parameters, and was trained on over 9 trillion nucleotides from more than 128,000 species across all domains of life. The model is based on the StripedHyena-2 architecture, processes context lengths of up to 1 million base pairs, and is Open Source (code, training data, and weights) available under Apache-2.0 license, including via GitHub, Hugging Face, and NVIDIA BioNeMo/NIM.
Fonctionnalités
| Deployment Model | Self-hosted (locally via GitHub/Vortex, NVIDIA NIM) or cloud-hosted (NVIDIA BioNeMo, hosted API) |
| Use Case Scope | Science & biotech: genome/DNA design, mutation prediction, drug discovery, agriculture, industrial biotechnology |
| Integrations | NVIDIA BioNeMo Framework, NVIDIA NIM microservice, Savanna training framework, Vortex inference code |
| License | Apache 2.0 (open model weights, code, training data) |
| Platform | GitHub, Hugging Face, NVIDIA BioNeMo/NIM, NVIDIA-hosted API, Together playground (Evo 1) |
| Price | Free/open source; NVIDIA-hosted API available as free trial (rate-limited) |
| Release Date | Evo 1: November 2024; Evo 2 preprint: February 19, 2025; Evo 2 in Nature: March 2026 |