Thermo Fisher Teams With NVIDIA and TetraScience to Bring AI Into the Lab

Thermo Fisher Scientific has announced a strategic collaboration with NVIDIA to integrate AI infrastructure across Thermo Fisher’s scientific instruments and laboratory systems. The companies aim to automate routine workflows and improve experiment design, data analysis, and instrument interaction using NVIDIA’s DGX Spark, NeMo, and BioNeMo platforms. No financial terms were disclosed. 

• DGX Spark is a compact desktop desktop AI supercomputer for running and fine-tuning large language and vision models locally, and can also act as an on-premise AI node or external accelerator for attached workstations.
• NVIDIA NeMo is a GPU-accelerated, open-source framework for building, fine-tuning, and deploying generative AI models, including large language models, multimodal models, and speech models.
• NVIDIA BioNeMo is a generative AI platform focused on biological sequences and structures, providing pretrained models and workflows for proteins, nucleic acids, and small molecules. It is designed for tasks such as protein property prediction, sequence design, molecular generation, and structure-aware modeling, and can run on NVIDIA GPUs on-premise or in the cloud via frameworks, blueprints, and NIM microservices.

In a separate agreement announced around the start of JPM 2026, TetraScience said it will integrate Thermo Fisher’s instruments and informatics software as part of a broader effort to support Scientific AI through interoperable, industry-aligned data infrastructure. The collaboration will focus on high-value scientific workflows at selected global biopharmaceutical organizations, including several that already deploy combined Thermo Fisher and TetraScience systems, providing an installed base for expanding AI-enabled use cases.

Thermo Fisher has been building a broader AI stack, including a 2025 collaboration with OpenAI to embed large language models across its PPD clinical research business and other operations.

The move also expands NVIDIA’s presence in scientific computing and life sciences R&D with a BioNeMo-centered push to make lab-in-the-loop “autonomous lab infrastructure” repeatable across partners. Also at JPM 2026, NVIDIA announced two major collaborations in biological AI—one with Eli Lilly to launch a $1 billion co-innovation lab in South San Francisco focused on integrating agentic wet labs with BioNeMo and Vera Rubin infrastructure, and another with Owkin to scale its OwkinZero reasoning model using NeMo RL and Nemotron for multimodal, agent-based biological modeling. We track developments like these in our weekly Substack newsletter Where Tech Meets Bio.

Topic: Industry Movers