Satellite and geo

Ai2 SatlasPretrain

Pretrained models from the Satlas project for Sentinel-2, Landsat and high-resolution aerial imagery. The predecessor of OlmoEarth, still used in TorchGeo.

The last open version came out in Jan 2024. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
Allen Institute for AI (Ai2), USA
First release
Aug 2023
Latest release
Jan 2024
Sizes
Swin-v2 and ResNet backbones (Base)
License
Commercial use allowedCode — Apache 2.0, weights — ODC-BY (attribution required)
Running
On your own serverNeeds a GPU
Industries
Science and research, Public sector, Manufacturing and logistics
Deploy on your server Hugging Face GitHub

What it does

  • Detecting objects in imagery: solar farms, wind turbines, ships
  • Mapping roads and buildings from aerial photos
  • A starting point for fine-tuning your own geo model

Where it is used

EnergyGeospatial analyticsMunicipal services

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. Пакет satlaspretrain_models
  2. SatlasPretrain (веса на HF)

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

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 Ai2 SatlasPretrain be used in a commercial project?

Yes. License: Code — Apache 2.0, weights — ODC-BY (attribution required). It allows commercial use, but it is still worth having a lawyer review the license before launch.

What hardware does Ai2 SatlasPretrain need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does Ai2 SatlasPretrain support Russian?

Language does not matter for this model: it does not work with text.

Where can I download Ai2 SatlasPretrain and what does it cost?

The Ai2 SatlasPretrain 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

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Satellite and geoAi2 OlmoEarthAllen Institute for AI (Ai2) · USACommercial use with conditions

Ai2'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 geoIBM–NASA PrithviIBM and NASA · USACommercial use allowed

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.

DetailsSatellite and geoIBM–ESA TerraMindIBM and the European Space Agency (ESA) · USA / EuropeCommercial use allowed

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).

Details

Source: github.com/allenai/satlaspretrain_models. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.