OPEN · TRUSTWORTHY · EXPAINABLE AI
Building AI for Climate Resilient Agriculture
AgriScienceFM is a European research initiative developing advanced AI systems that understand the interactions between plant phenotypes, environment, and agricultural management. By combining satellite imagery, climate data, agronomy knowledge, and plant phenotyping, the project aims to accelerate scientific discovery and support climate-resilient food systems
3
Foundation models
4
Scientific use cases
FAIR
Open datasets
EU
Horizon Europe
Key Highlights
What sets AgriScienceFM apart
Combining satellite imagery, climate data, agronomy knowledge and plant phenotyping into reusable, open agriculture foundation models
01
Open agricultural foundation models
Reusable AI foundations, publicly accessible to the research community
02
AI for climate-resilient agriculture
Accelerating innovation for sustainable, climate-adapted food systems
03
Multimodal learning
Across genetics, environment, and management, in one framework
04
FAIR and reusable datasets
Findable, accessible, interoperable and reusable by design
05
Benchmarking suite
Standardized evaluation for agricultural AI performance
06
Explainable & trustworthy AI
Transparent, reliable systems built on responsible AI practices
What is AgriScienceFM?
One of the most complex sciences in the world
Crop growth and food production depend on the interaction between genetics, environment, and management. AgriScienceFM develops a new generation of foundation models capable of learning from large-scale multimodal data, creating reusable AI foundations that adapt to many agricultural challenges with less training data and greater reliability
G
Genetics
E
Environment
M
Management
Learning from large-scale multimodal data
DATA SOURCES
- Earth Observation & satellite imagery
- Weather & climate data
- Soil information
- Agronomy documents & advisory materials
- Plant imaging & phenotyping datasets
The Three AgriFMs
Three foundation models
ENVIRONMENT
AgriFM-E
A spatio-temporal model trained on satellite imagery, weather, soil and crop phenology
MANAGEMENT
AgriFM-M
A document-understanding model trained on agronomy literature, advisory and extension materials
GENETICS & GROWTH
AgriFM-G
A plant-imaging model trained on multimodal phenotyping and imaging datasets
Our Vision
A future where open, trustworthy AI tools improve food security, sustainability and climate resilience
Accessible to agricultural scientists, farmers, policymakers and innovators alike, built on open science and responsible AI
Explore the science behind AgriScienceFM
Dive into the project, the use cases, and the open deliverables shaping the future of agricultural AI
Committed to open science, interdisciplinary collaboration, and responsible AI innovation for sustainable agriculture
FUNDING
Funded under the Horizon Europe programme
© 2026 AgriScienceFM · Privacy & Cookies
Open · Trustworthy · Explainable