Cambridge University’s Energy & Environment Group Deploys Planetary-Scale Foundation Model on Vultr Cloud GPU Infrastructure Powered by AMD Instinct™ MI325X GPUs

Vultr, the world’s largest privately-held cloud infrastructure company, and AMD, a leader in high-performance and AI computing, have announced their support for the University of Cambridge’s Energy & Environment Group in developing TESSERA, a groundbreaking AI foundation model designed to monitor environmental change internationally at unprecedented scale.

Vultr is providing the computational infrastructure, based on AMD Instinct MI325X accelerators, necessary to generate global embeddings covering Earth’s land surface from 2017 to 2025, enabling applications in agriculture, biodiversity conservation, and renewable energy infrastructure monitoring. The collaboration represents a significant advancement in applying artificial intelligence to planetary conservation challenges.

TESSERA processes data from European Space Agency Sentinel satellites to create detailed embeddings – compressed AI representations – of every 10-meter square of Earth’s surface. TESSERA represents the most accurate pixelwise foundation model ever created from Sentinel satellite data, and leads the field in results and reproducibility. These embeddings enable researchers, governments, and organizations worldwide to track environmental changes, from crop health monitoring to deforestation detection, at a scale and accuracy previously unattainable.

“This collaboration positions Vultr at the intersection of cutting-edge AI infrastructure and planetary conservation,” said Kevin Cochrane, Chief Marketing Officer at Vultr. “By providing the computational resources required for TESSERA and embracing open source technology, Vultr is enabling research that has immediate real-world impact, from helping smallholder farmers optimize crop yields to tracking biodiversity loss in critical ecosystems.”

Stephanie Dismore, Senior Vice President, EMEA at AMD added, “TESSERA shows what becomes possible when world-class research teams can scale AI with the right compute and an open software foundation. By running on Vultr cloud infrastructure powered by AMD Instinct™ MI325X GPUs and the AMD ROCm™ open software stack, University of Cambridge researchers can efficiently process petabytes of satellite data and generate planetary-scale embeddings that help turn environmental observation into actionable insight.”

The project will leverage Vultr’s AMD-powered infrastructure, including 4×8 MI325X GPU configurations, to process and host the embeddings. This infrastructure will support both the intensive training requirements and long-term storage needs of the global dataset.

Professor Anil Madhavapeddy, Professor of Planetary Computing at Cambridge’s Department of Computer Science & Technology and Co-Director of the Cambridge Centre for Earth Observation, emphasized the project’s commitment to open access: “Our goal is to democratize access to planetary-scale environmental monitoring. By making TESSERA’s embeddings freely available under a CC-BY license and publishing the complete training pipeline, we’re ensuring that any researcher, government, or organization worldwide can deploy this technology for their specific conservation needs.”

The Energy & Environment Group will release not only the embeddings but also a fully reproducible pipeline for training the model from scratch on Vultr infrastructure, enabling deployment and fine-tuning for region-specific applications.

The project’s computational requirements include both GPU resources for model training and CPU capacity for processing raw satellite observations.

Three Key Application Verticals

The collaboration will develop case studies across three primary verticals:

• Agriculture & Farming Optimization: Enabling precision agriculture through crop health monitoring, yield prediction, and risk management at 10-meter resolution, with particular focus on supporting smallholder farmers in developing regions.
• Biodiversity Monitoring & Conservation: Tracking habitat changes, species distribution, and ecosystem health across diverse environments, from tropical forests to UK hedgerow networks critical for species like hedgehogs.
• Renewable Energy Infrastructure: Detecting and mapping solar installations, wind farms, and other renewable energy infrastructure to support energy transition planning and policy development.

“This partnership exemplifies how cloud infrastructure can accelerate scientific research with genuine planetary impact,” concluded Professor Madhavapeddy. “We’re enabling a new paradigm for open, accessible environmental monitoring that can inform policy decisions and conservation actions in the UK and worldwide.”

The collaboration began in November 2025, with the first global embeddings now available for public access since early 2026.

Source:BusinessWire