NVIDIA today announced the NVIDIA BioNeMo Agent Toolkit, which provides domain-specific tools and skills for the life sciences agent era.
Including more than a decade of NVIDIA libraries, tools, and open models, the toolkit enables AI agents, scientists, and labs to collaborate by collecting data, reasoning about findings, running computational experiments, and suggesting next best steps to accelerate discovery.
It provides any AI agent or platform – from general-purpose assistants to specialized scientific agents, software platforms, and biopharmaceutical internal systems – with the tools needed to synthesize and summarize scientific knowledge, call models, evaluate results, reason, and take next actions.
The toolkit includes NVIDIA BioNeMo™ and powered by NVIDIA NIM™ microservices, NVIDIA Parabricks®NVIDIA NeMo™ and NVIDIA Nemotron™ technologies, along with accelerated computing and skills – providing an open and reliable foundation for the life sciences.
More than 50 leading companies already use it to advance scientific discovery, leveraging skills that agents can call on for tasks such as protein structure prediction, molecular docking, genetic chemistry, genomic analysis, protein design and biomarker discovery.
Boundary models are the brains. BioNeMo is the scientific toolbox. Together, they give AI agents the skills of a PhD research assistant and the speed of a supercomputer, For the first time, researchers can create AI agents that understand scientific knowledge, use scientific tools, and perform scientific workflows. This is a new way of doing science – one that can dramatically accelerate discovery across biology, chemistry, genomics and medicine.”
Jensen Huang, Founder and CEO of NVIDIA
Open modeling and research organizations – including the Arc Institute, the Open Molecular Software Foundation, and the University of Washington’s Institute for Protein Design (IPD) – are working with NVIDIA to use BioNeMo to advance frontier models and make them more accessible through agent-ready workflows. The IPD collaboration has accelerated runtimes for state-of-the-art biodesign models such as RosettaFold3, resulting in 2x faster throughput than the previous generation model, and many additional applications to accelerate protein design efforts are underway, giving researchers tools at a scale and cost not possible before.
“Every tool we’ve built for protein design is only as powerful as the scientists who can effectively access it.” said David Baker, professor of biochemistry at the University of Washington School of Medicine and director of the Institute for Protein Design. “The next leap in science will not come from a single discovery; it will come from the speed of iterative designs and agents that can repeatedly reason through the complexity of biology at a speed that humans never could.”
Agent-ready tools and skills for the life sciences
The life sciences are one of the world’s most important scientific frontiers, with global scientific R&D reaching $3.8 trillion and annual pharmaceutical budgets approaching $300 billion.
Agent workflows can help the industry iterate faster while reducing costs and maximizing the likelihood of success. With the toolkit allowing developers to turn general-purpose agents into life science agents in minutes, researchers can run experiments faster, continuously learn from results, and close the loop between hypothesis and discovery, with some companies extending this iteration to physical labs.
A general-purpose agent may struggle to effectively navigate scientific workflows, needing to infer the right tools, inputs, outputs, and biological meaning along the way. With the BioNeMo Agent Toolkit, agents can call the right tools, interpret results more accurately, and gain scientific knowledge faster and more reliably.
NVIDIA optimizes the entire BioNeMo platform by turning libraries, models, and frameworks into tools called by agents.
This includes utilization NVIDIA Agent Toolkit technologies such as NVIDIA Nemotron open models for reasoning, NVIDIA NeMo RL library for reinforcement learning and NVIDIA NemoClaw™ blueprints for secure, private agents that can reason about tasks, call tools, and continuously interact with data.
NVIDIA NIM microservices help agents call models and perform tasks. NVIDIA OpenShell™ The runtime provides a controlled executable environment.
Toolbox components allow agents to complete workflows such as:
- Virtual View: Agents can help researchers identify small-molecule drug candidates by creating and screening compounds, binding them to a target, predicting binding potency, and filtering for drug-like properties. The agent can then provide which candidates should be prioritized, compressing screening timelines from days to minutes.
- Genomic Analysis and Target Discovery: Factors can help researchers turn raw sequence data into prioritized genetic insights and biological targets. NVIDIA Parabricks accelerates alignment and variant calling, while genome-wide foundational models score variant results and the factor ranks the most disease-relevant candidates for further study.
- Protein Binder Design: Agents can help researchers design and computationally validate candidates before work begins, compressing traditionally labor-intensive design work.
- Deep biomedical research: Agents connect real-world data with reasoning models to improve the efficiency and accuracy of various scientific and clinical development processes, including literature review, protocol generation, clinical trial review, and pharmacovigilance with the NVIDIA Biomedical AI-Q Research Agent.
- Medical imaging analysis: Agents can help researchers process, segment, synthesize, and reason medical imaging data to support biomarker discovery, accelerating evidence generation across research workflows.
The life sciences ecosystem is built with NVIDIA BioNeMo
Companies across the tech and life sciences ecosystem are using the toolbox to advance agent workflows.
Frontier Labs and scientific agent manufacturers, including Anthropic, Edison Scientific, Lila Sciences, OpenAI and Owkin, are integrating with BioNeMo to help agents move from answering questions to completing scientific work. NVIDIA’s accelerated models and analysis libraries help reduce time from hypothesis to knowledge.
Scientific data and workflow platforms from Benchling, Certara, Databricks, Snowflake and Seqera use the BioNeMo Agent Toolkit to connect data systems with AI-powered science. BioNeMo’s skills can help agents search biological and chemical datasets, prepare model-ready inputs, launch reproducible workflows, analyze results, and feed information back directly to the platforms scientists and data teams already use every day.
Diagnostic and pharmaceutical companies, including Lilly and Natera, use the BioNeMo Agent Toolkit to scale repeatable agent workflows in discovery, translational research and clinical imaging.
Native AI biology companies, including Boltz, Basecamp Research, Chai Discovery, Dyno, PerturbAI, and Proxima, have partnered with NVIDIA to develop tools to accelerate model-based therapeutic design workflows.
Computer-aided drug discovery software providers, including Dassault Systèmes, Cadence (OpenEye), and Schrödinger, are incorporating the toolkit’s capabilities into scientific applications used by discovery teams. Agents can then help orchestrate molecular production, connection, and prediction, turning computer-aided design platforms into systems where researchers can ask questions, initiate analyses, and identify next best actions more quickly.
Laboratory instrumentation and automation companies, including Automata, HighRes, Tecan, Thermo Fisher and autonomous data generation platform Medra, are connecting systems with computational discovery powered by BioNeMo skills.
AI cloud and AI infrastructure companies including Baseten, Modal and Nebius use the toolkit to help developers build life science workflows as trusted hosted services. By supporting BioNeMo’s skills and tools through extensible APIs, managed computational inference environments, and production, these companies can help move agent biology workflows from prototypes to accessible services for researchers and businesses.
Availability
The BioNeMo Agent Toolkit and skills are available now through the NVIDIA Developer Resources page and GitHub.
