NIM for Evo 2

Type: Microservice Tags: NVIDIA, NIM, BioNeMo, Evo 2, DNA, genome, biological foundation model, synthetic biology, drug discovery Related: NVIDIA-BioNeMo, NVIDIA-NIM, NIM-for-Boltz2, NIM-for-OpenFold3, NIM-for-AlphaFold2, NIM-for-ProteinMPNN, NIM-for-RFdiffusion, NVIDIA-Clara, NVIDIA-AI-Enterprise, NGC Sources: https://docs.nvidia.com/nim/bionemo/evo2/latest/overview.html; https://docs.nvidia.com/nim/bionemo/evo2/latest/index.html Last Updated: 2026-04-29

Summary

NIM for Evo 2 is NVIDIA’s NIM microservice for the Evo 2 biological foundation model. Evo 2 interprets and generates DNA sequences across biological scales, from individual molecules to entire genomes, with sensitivity to single-nucleotide changes for zero-shot predictions and biological system design.

Detail

The current docs describe Evo 2 as a biological foundation model trained by Arc Institute. NVIDIA’s NIM packaging exposes Evo 2 for DNA sequence generation, model forward-pass outputs, benchmarking, model selection, and custom or fine-tuned model checkpoints.

Evo 2 belongs with BioNeMo NIMs, but its query intent is different from structure-prediction pages. NIM-for-AlphaFold2, NIM-for-OpenFold3, and NIM-for-Boltz2 answer structure and interaction questions; Evo 2 answers DNA sequence interpretation/generation, mutation-effect prediction, CRISPR-style design, and synthetic biology questions.

Connections

• NVIDIA-BioNeMo - BioNeMo includes Evo 2 in its biomolecular NIM ecosystem.
• NVIDIA-NIM - delivered as an NVIDIA NIM microservice.
• NIM-for-Boltz2 - structure/binding-affinity prediction adjacent to drug discovery workflows.
• NIM-for-OpenFold3 - biomolecular complex prediction adjacent to sequence-to-structure pipelines.
• NIM-for-AlphaFold2 - protein structure prediction NIM in the same BioNeMo family.
• NIM-for-ProteinMPNN and NIM-for-RFdiffusion - protein design NIMs adjacent to biological sequence and design workflows.
• NVIDIA-Clara - healthcare/life-sciences umbrella platform.
• NVIDIA-AI-Enterprise - enterprise support and production deployment context.
• NGC - distribution path for NIM containers and model assets.

Source Excerpts

• “interpret and generate DNA sequences”
• “entire genomes”
• “zero-shot predictions”