Key points
- NVIDIA released over 1,400 open source projects and 250 models to support scientific research.
- The company launched free, modular courseware for accelerated computing and deep learning.
- Partnerships with national AI factories aim to provide sovereign compute access for startups.
The short version
- NVIDIA announced expanded access to open models, datasets and software for European scientists.
- New free training materials from the Deep Learning Institute remove enrollment and fee barriers.
- Collaborations with cloud partners and AI factories in France, Spain and Sweden support startups.
The details
NVIDIA announced at the AI in Science Summit in Dublin that it is expanding access to advanced technology for Europe’s scientific community. The company released more than 250 open models and 250 open datasets alongside over 1,400 open source software projects. These resources cover large language models, healthcare, robotics, climate science and physical AI. According to the NVIDIA blog, these materials are available on Hugging Face and GitHub to help researchers assemble tools and build expertise.
The pledge includes scientific software such as the BioNeMo Agent Toolkit, PhysicsNeMo libraries and Earth-2 models. These tools aim to accelerate research in biology, chemistry, molecular design and material sciences. NVIDIA also introduced the Nemotron family of architectures, which provides open weights and training recipes for agentic reasoning capabilities. This open ecosystem is designed to help smaller labs and early-career scientists overcome barriers to entry.
The context
The European Commission launched the AI in Science Enablers Initiative during the summit. This framework connects private-sector support to Europe’s research community under the RAISE resource. European Commissioner Ekaterina Zaharieva stated that bringing together research, industry and partners is essential for turning AI ambition into reality. The initiative focuses on providing compute, cloud resources, data and collaborative project support for researchers and talent development.
NVIDIA is collaborating with cloud infrastructure partners and national AI factories in France, Spain and Sweden. These partnerships aim to open routes for startups and small- and medium-sized enterprises to access sovereign European compute. The NVIDIA Inception program provides technical support and compute resources to lower barriers for startups. AI2F in France has already collaborated with the program for four months, showing strong results.
What to watch
- The BSC AI Factory in Spain is rolling out its first full accelerator program after a successful pilot.
- Other European national AI factories have expressed interest in replicating this startup support model in 2027.
- The NVIDIA Deep Learning Institute has launched an open source courseware hub with freely downloadable materials.
- These courses are modular and can be adapted for university curricula or developer community events without fees.
What it means for organisations
Organisations should assess whether their research or development teams can leverage these open resources to reduce costs. IT leaders need to verify if existing infrastructure supports the new models and datasets. Procurement teams should evaluate partnerships with national AI factories for sovereign compute access. Training departments can integrate the free courseware into internal upskilling programs. Ensure that data governance policies align with the use of open source scientific tools and external compute services.
What to do and how to stay safe: NVIDIA
- Review internal policies to determine if teams can use open source models from Hugging Face or GitHub.
- Check whether current hardware and cloud contracts support the computational requirements of new scientific tools.
- Identify local national AI factories or cloud partners that offer sovereign compute access for research projects.
- Evaluate the free NVIDIA Deep Learning Institute courseware for potential integration into staff training programs.
Step-by-step guide: Secure Home Routers for Small Business Networks
General security guidance from the TechSec Headlines newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
What resources has NVIDIA released for European scientists?
NVIDIA released over 1,400 open source projects, 250 open models and 250 open datasets on Hugging Face and GitHub.
How can startups access compute resources through this initiative?
Startups can use the NVIDIA Inception program and collaborate with national AI factories in France, Spain and Sweden.
Are there fees for the new training materials?
No, the NVIDIA Deep Learning Institute courseware is freely downloadable, openly licensed and has no enrollment fees.
