NVIDIA BioNeMo Integrates With Anthropic Claude Science

Tools & Releases

NVIDIA BioNeMo Integrates With Anthropic Claude Science

NVIDIA’s BioNeMo Agent Toolkit has integrated into Anthropic’s new Claude Science workbench to execute GPU-accelerated biology and chemistry workflows.

AZAli Zayed · Founder & EditorJuly 1, 20262 min read✓ Independently fact-checked
The quick version
  • Anthropic’s Claude Science workbench now integrates the NVIDIA BioNeMo Agent Toolkit to run accelerated life sciences workflows.
  • Researchers can use natural language to command AI agents to perform complex tasks like protein structure prediction and genomic analysis.
  • The integration connects Claude’s agents to specialized NVIDIA-accelerated models including Evo 2, Boltz-2, and OpenFold3.

Anthropic has launched Claude Science, an AI-powered workbench designed for scientific research that integrates directly with the NVIDIA BioNeMo Agent Toolkit. According to an announcement from NVIDIA, the integration allows life sciences researchers to build and run complex, GPU-accelerated computational workflows using natural language commands instead of manual coding or environment configuration.

The system works by connecting Anthropic’s conversational agents to NVIDIA’s suite of specialized scientific models and microservices, including Evo 2, Boltz-2, and OpenFold3. When a scientist describes a task in plain English—such as predicting a protein structure or analyzing a genomic sequence—Claude Science interprets the intent. It then uses the BioNeMo Agent Toolkit to select the correct computational tool, format the inputs, and execute the job on NVIDIA’s accelerated computing infrastructure.

Why it matters

AI agents are only as fast as the tools they orchestrate. In the life sciences, tasks like molecular design, genomic analysis, and clinical research require massive computational scale. By linking Anthropic’s reasoning models with hardware-accelerated microservices, this partnership aims to eliminate the manual software bottlenecks that typically slow down digital biology research. According to NVIDIA, 18 of the top 20 pharmaceutical companies already deploy BioNeMo technology, indicating a highly receptive enterprise market for these agentic workflows.

This integration also highlights a broader shift in the AI landscape. As developers push conversational models into highly specialized, domain-specific roles, general-purpose assistants are evolving into functional operating systems for industry. For organizations evaluating how these enterprise-grade platforms compare to standard consumer assistants, keeping track of the best AI chatbots remains essential as the technology bifurcates into general and scientific applications.

What it means for researchers

For research teams, the primary benefit is the reduction of friction in the scientific feedback loop. Instead of writing custom scripts to bridge different structural biology and chemistry tools, scientists can run end-to-end pipelines—such as generating and validating potential cancer inhibitors—from a single interface. The BioNeMo Agent Toolkit provides the necessary context for Claude’s agents to understand the purpose and required inputs of each scientific capability, allowing researchers to inspect outputs and refine their queries iteratively without leaving the workbench.

18 of 20top pharmaceutical companies using NVIDIA BioNeMo

Frequently asked questions

What is Claude Science?

Claude Science is an AI workbench developed by Anthropic that allows researchers to use natural language agents to build, orchestrate, and execute scientific workflows.

How does the BioNeMo Agent Toolkit work inside Claude Science?

The toolkit packages NVIDIA’s accelerated scientific models as callable skills, enabling Claude Science’s agents to select the correct tools, format inputs, and run workflows on GPU resources.

Which scientific models can be accessed through this integration?

The integration provides direct access to specialized models and accelerated workflows, including Evo 2, Boltz-2, and OpenFold3.

Our tested pick

Explore our hands-on review of the best AI chatbots to see how the top models compare in performance and features.

Best AI Chatbots in 2026 (Tested & Ranked) →

Source: NVIDIA. Published July 1, 2026.

AZ
Ali Zayed
Founder & Editor · AI Tools Worth

Ali has hands-on tested 50+ AI tools and tracks model releases daily. Every verdict here comes from real, paid usage — never vendor demos or sponsored placements.

AI Tools Worth is independent and unsponsored. Some linked guides contain affiliate links — they never change our verdicts.