geospatial
AlphaEarth Foundations (Satellite Embedding v1 Annual)
Google DeepMind geospatial embedding model; about 23.9% lower average error than next-best across 15 mapping tasks (vs CCDC, SatCLIP, Prithvi, Clay), ET R2 0.58.
Levelmodel
Classgeospatial
Parametersnot in the notes
Contextnot in the notes
ReleasedJul 2025
Opennessproprietary
Runs onnot in the notes
Statusactive
R&D Topics
Identity
- Developer
- Google DeepMind
- Family
- not in the notes
- Released
- Jul 2025 sourcesinferred from page
- Weights published
- –
- Record checked
- 10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- other Embedding field model; 64-dim embedding per 10 m pixel
- Total parameters
- – sources
- Active parameters
- –
- Checkpoint parameters
- –
- Context length
- not in the notes
- Input
- image
- Output
- image
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- proprietary
- Licence
- not in the notes sources
- Access routes
- cloud marketplace
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- not in the notes
- Footprint
- Consumed via Earth Engine (cloud); embeddings published for 2017 to 2024
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Benchmark Results
The notes record no benchmark result for this model.
Lineage
The notes record no lineage for this model.
Industry Use Cases
No industry use case links this model.
Links
ModelAlphaEarth Foundations (Satellite Embedding v1 Annual)Google DeepMind, geospatial
AlphaEarth Foundations (Satellite Embedding v1 Annual)Google DeepMind, geospatialR&D TopicsEarth Observation Foundation ModelsPlantation and Natural Forest Mapping from Satellite and Drone Imagery
Use CasesSatellite Monitoring (Land Cover, Floods, Forests)Public SectorLand Use and Crop MappingAgricultureDeforestation Baselines and Supply-Chain Monitoring (EUDR)AgricultureOil Palm Yield PredictionAgriculture+1 moreUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Astra Agro Lestari TbkIn production, 21 Sep 2026PT LAT Trisakti (no named client)Planned, 12 Aug 2026
PTPN IV PalmCo (Regional 1 to 7 staff)Planned, 22 Jul 2026
BRIN Data Science and Information Research Center (PRSDI)Research, 10 Mar 2026+7 moreDeployments of these use cases; none names this model.
PT Astra Agro Lestari TbkIn production, 21 Sep 2026PT LAT Trisakti (no named client)Planned, 12 Aug 2026PTPN IV PalmCo (Regional 1 to 7 staff)Planned, 22 Jul 2026BRIN Data Science and Information Research Center (PRSDI)Research, 10 Mar 2026+7 moreDeployments of these use cases; none names this model.Organisations 1
R&D topics 2
Earth Observation Foundation ModelsModels pretrained on large satellite and aerial archives that are fine-tuned for land cov…exemplarPlantation and Natural Forest Mapping from Satellite and Drone ImageryMapping tree crops such as oil palm and separating natural forest from plantations, for y…used by
Sources
- AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data arxiviq.substack.com https://arxiviq.substack.com/p/alphaearth-foundations-an-embedding
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "AlphaEarth Foundations (Satellite Embedding v1 Annual)", https://aurion.semai.id/models/google-deepmind-alphaearth-foundations/, records of October 2026.