AI Potluck
Back to Gap Map Model components / Scientific AI models

Prithvi-EO

NASA IMPACT
open weights / Overall score: 3.1

Prithvi-EO is an Earth-observation foundation-model family jointly developed by NASA and IBM. Its spatiotemporal vision transformer consumes Harmonized Landsat and Sentinel imagery together with location and acquisition-time metadata, and the release includes several model sizes plus configurations and notebooks for downstream remote-sensing tasks.

Verified 2026-09-10 via the Hugging Face model card and NASA IMPACT repository.

Openness

3 high confidence
3.0
license
Apache-2.0(OSI)
weights
open(Apache-2.0 ungated checkpoint)
data
documented-not-released(4.2M HLS V2 samples and six bands described
code
partial(inference and downstream fine-tuning code, not the pretraining pipeline)

The weights are permissively licensed and the training inputs are described, but the repository starts at inference and downstream fine-tuning rather than reproducing pretraining.

Adoption

1 high confidence
1.0

1,899 downloads in the trailing 30 days for the declared 600M checkpoint.

Capability

4 high confidence
4.0

Three of four attributes place it one rung below TerraMind; masked pretraining reconstruction does not by itself count as a scientifically meaningful generative output.

Verified 2026-09-10