OlmoEarth v1.3: State-of-the-art embeddings for the whole planet

Summary

OlmoEarth v1.3 is a satellite embedding model that reads a year of radar and multispectral time series and outputs a 128-dimensional vector per 10-meter pixel, with the first 64 dimensions usable as a compact variant. It scores highest on 20 of 27 evaluations across six continents, outperforming AEF and Tessera v2, and generalizes best on the new PLANTEUR crop-type benchmark. Precomputed global embeddings at 10-meter resolution for 2017–2025 covering land and coastal waters, along with model weights, data, and code, are fully open on Source Cooperative, enabling anyone to track change or compare places without a GPU. Training used 53% of v1.2’s compute and 18% of v1’s, with 50% and 22% of their carbon emissions, while matching v1.2’s fine-tuning performance. The OlmoEarth Embeddings Explorer provides a browser-based false-color map to compare years, find similar locations, and train classifiers on the spot without an account.

Takeaways

1. 1

   OlmoEarth v1.3 embeddings score highest on 20 of 27 evaluations across six continents, outperforming AEF and Tessera v2, including on the new PLANTEUR crop-type benchmark.

2. 2

   Precomputed global embeddings at 10-meter resolution (64- and 128-dimension) are openly available on Source Cooperative, enabling change detection and similar-place search without GPU or account.

3. 3

   Training OlmoEarth v1.3 required 53% of v1.2’s compute time and 18% of v1’s, with 50% and 22% of the carbon emissions respectively, while maintaining strong fine-tuning performance.

FAQ