At a glance: Claude Science invokes NVIDIA-accelerated life sciences tools through natural language agents, enabling complex analyses such as protein structure prediction and drug optimization without manual configuration.
Anthropic has unveiled Claude Science, an AI workbench for scientific research. NVIDIA integrates its BioNeMo Agent Toolkit as a resource to provide researchers with GPU-accelerated computing in genomics, proteomics, and drug design directly within the workflow.
Anthropic announced Claude Science this week, an AI workbench through which scientists can communicate with agents in natural language to execute their workflows entirely end-to-end. NVIDIA has anchored its BioNeMo Agent Toolkit as an integrated resource in Claude Science. The toolkit provides NVIDIA-accelerated functions as callable skills, enabling Claude agents to automatically select the appropriate tool, prepare valid inputs, and execute the workflow — all with access to NVIDIA compute resources that can be deployed anywhere.
The integration addresses tasks in genomics, proteomics, single-cell analysis, and clinical research. A researcher describes a task in natural language — such as analyzing a genomic sequence, predicting a protein structure, or designing a potential inhibitor. Claude Science interprets the request and coordinates work across pre-configured, domain-specialized agents. The BioNeMo Agent Toolkit supplies these agents with the necessary context and information about the purpose and required inputs of each skill. Scientists can review results, refine their questions, and determine next steps without losing focus on the science.
A concrete use case is the generation of better inhibitors for common cancer targets. A researcher begins with a known carcinogenic antigen mutation and asks Claude to design numerous potential inhibitors. Claude Science then leverages high-throughput prediction, optimization, and validation of inhibitors through the BioNeMo Agent Toolkit and NVIDIA NIM microservices.
At its core, the toolkit builds on accelerated scientific components: NVIDIA Parabricks reduces genomic analysis from hours to minutes. RAPIDS-singlecell, developed by scverse, reduces preprocessing and clustering of 1.3 million cells from 52 minutes to 25 seconds. This acceleration enables autonomous agents to make decisions at scientific speed. Large pharmaceutical companies already use NVIDIA technologies in drug discovery, genomics, medical imaging, and protein engineering. 18 of the 20 largest pharmaceutical companies use BioNeMo.
Source: blogs.nvidia.com · Published June 30, 2026
Lumi AI News — AI-assisted curation in accordance with Article 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.7.2.