Sign in
← AI PulseAug 31, 2026

Deep · news · Single-source brief

NVIDIA BioNeMo NIM Microservices Integrate with Claude Science for Protein Structure Prediction

NVIDIA BioNeMo Agent Toolkit, integrated with Claude Science and NVIDIA NIM microservices, enables AI agents to orchestrate protein structure prediction workflows.

By Illumora Editorial

Source · Aug 31, 2026, 4:30 PM · On Illumora · Aug 31, 2026, 4:37 PM

Media from the primary source — shown here so you can stay on Illumora.

Rewritten from one allowlisted primary — not independent enterprise reporting. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →NVIDIA Developer Blog — Run NVIDIA BioNeMo NIM Microservices for Protein Structure Prediction in Claude Science | NVIDIA Technical Blog
Save

NVIDIA has integrated its BioNeMo Agent Toolkit with Anthropic's Claude Science and NVIDIA NIM microservices. This integration allows AI agents to manage protein structure prediction workflows, utilizing multiple-sequence alignment and various folding models.

The BioNeMo Agent Toolkit packages over a decade of BioNeMo life sciences models, libraries, and workflows into agent-callable skills for biology, chemistry, genomics, and drug discovery. This toolkit is designed to run with any agent framework and aims to bridge the gap in orchestrating domain-specific scientific tools.

Key Points

  • The NVIDIA BioNeMo Agent Toolkit integrates with Claude Science and NVIDIA NIM microservices.
  • This integration facilitates AI agents in orchestrating protein structure prediction workflows.
  • The toolkit includes BioNeMo life sciences models, libraries, and workflows as agent-callable skills.
  • It supports applications in biology, chemistry, genomics, and drug discovery.
  • On internal benchmarks, BioNeMo skills increased task correctness from 60% to 100% and approximately doubled token efficiency.
  • The workflow requires a workstation or cloud machine with an NVIDIA L40S GPU or NVIDIA H100 GPU and Claude Science installed.
  • The setup needs about 700 GB of storage for the workflow, including a 490 GB UniRef30 database for msa-search NIM and 30–40 GB for Boltz-2 and OpenFold3 containers.

Context

According to the NVIDIA Developer Blog, agentic AI is transforming research by allowing AI scientists to read papers, propose hypotheses, call models, and prioritize experiments. While coding agents have proven valuable in software engineering, scientific research presents more demanding and iterative challenges, often requiring domain-specific tools with varied environment requirements or APIs. The NVIDIA BioNeMo Agent Toolkit addresses this by providing specialized domain expertise to enable complex scientific workflows.

NVIDIA and Anthropic collaborated to integrate the BioNeMo Agent Toolkit into Claude Science, allowing agents to discover, launch, and call BioNeMo NIM microservices directly. Claude Science can run in various configurations, including on a machine with a GPU, or connecting to a remote GPU via SSH, HPC, or cloud computing on Modal. The tutorial focuses on running the platform on a machine with an NVIDIA L40S GPU or NVIDIA H100 GPU.

Why It Matters

This integration provides researchers and developers with a structured approach to automate complex scientific workflows, particularly in protein structure prediction. The ability to orchestrate specialized models and tools through an agent framework can accelerate hypothesis testing and experimental design in life sciences.

What To Do

  • Explore the BioNeMo Agent Toolkit on GitHub to apply the workflow to other monomers and complexes.
  • Note the hardware requirements, including an NVIDIA L40S GPU or NVIDIA H100 GPU and 700 GB of storage, for setting up the workflow.
  • Review the Claude Science product documentation for GPU compatibility information.
  • Consider the steps for configuring Claude Science to expose local or remote GPU resources for running BioNeMo NIM microservices.