IBM–NASA Prithvi
Foundation models for Landsat and Sentinel-2 satellite imagery that account for image time series. Ready-made fine-tunes for floods, burn scars and crop types, plus a separate weather model, WxC.
- Developer
- IBM and NASA, USA
- First release
- Jul 2023
- Latest release
- Sep 2025
- Sizes
- tiny – 600M (imagery), 2.3B (Prithvi WxC weather)
- License
- Commercial use allowedApache 2.0 (Prithvi-EO), CDLA-Permissive 2.0 (Prithvi WxC)
- Running
- On your own serverAlso runs without a GPU
- Industries
- Manufacturing and logistics, Finance, Science and research, Public sector
What it does
- Mapping crops and field condition over the season
- Assessing flood zones and burn scars after natural disasters
- Monitoring changes in buildings and land use
- Fine-tuning for your own object classes on satellite data
Where it is used
Hardware requirements
Versions
- Prithvi-EO 2.0 tiny и 100M
- Prithvi-EO 2.0 300M и 600M
- Prithvi WxC 2.3B
- Prithvi-EO 1.0 100M
How to run it
I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.
Frequently asked questions
Can IBM–NASA Prithvi be used in a commercial project?
Yes. License: Apache 2.0 (Prithvi-EO), CDLA-Permissive 2.0 (Prithvi WxC). It allows commercial use, but it is still worth having a lawyer review the license before launch.
What hardware does IBM–NASA Prithvi need?
At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.
Does IBM–NASA Prithvi support Russian?
Language does not matter for this model: it does not work with text.
Where can I download IBM–NASA Prithvi and what does it cost?
The IBM–NASA Prithvi weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.
How I deploy it for clients
- SelectionI pick the model size for your task and hardware and test it on your examples.
- DeploymentI deploy it on your server or in a closed network and provide an API.
- Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
- IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.
Similar models
A multimodal Earth model: understands optical and radar imagery, terrain, vegetation index and land use maps, and can generate a missing data type (for example, a "see-through-clouds" image from radar).
DetailsSatellite and geoAi2 OlmoEarthAllen Institute for AI (Ai2) · USACommercial use with conditionsAi2's family of models for Sentinel-1, Sentinel-2 and Landsat imagery, with ready-made fine-tunes for mangroves, deforestation and ecosystem types. The license excludes the extractive industries.
DetailsSatellite and geoAi2 SatlasPretrainAllen Institute for AI (Ai2) · USACommercial use allowedPretrained models from the Satlas project for Sentinel-2, Landsat and high-resolution aerial imagery. The predecessor of OlmoEarth, still used in TorchGeo.
DetailsSource: huggingface.co/api/models?author=ibm-nasa-geospatial. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.


