OlmoEarth v1.3: State-of-the-art embeddings for the whole planet
- ID
- 38347
- Status
- summarized
- Published
- 23 Oct 2026, 12:11 AM
- Fetched
- 23 Oct 2026, 1:21 AM
- Provider
- Hugging Face Blog
- Category
- developer-ai
- Original URL
- https://huggingface.co/blog/allenai/olmoearth-v1-3
- Source URL
- https://huggingface.co/blog/feed.xml
Summary
- Score
- 6.0
- Created
- 23 Oct 2026, 1:21 AM
- Tags
- Audience
- developersai_ml_learnerssaas_startup_founders
What happened
Ai2 released OlmoEarth v1.3, a satellite embedding model that turns a year of Sentinel-1 radar plus Sentinel-2 and Landsat multispectral time series into one 128-dimensional vector per 10-meter pixel, with a 64-dimensional version also usable. Ai2 precomputed annual embeddings for the entire planet and published them on Source Cooperative, alongside a no-sign-in browser Embeddings Explorer for exploring and customizing maps. Across 27 evaluations on six continents against AlphaEarth Foundations and Tessera v2, the 128-dim embeddings ranked highest on 20 of 27 evals and the 64-dim version on 18 of 27; the release also introduces PLANTEUR, a tropical crop-type benchmark.
Why it matters
If you work on agriculture, land-cover, flood, or urban-change mapping in Malaysia or Southeast Asia, the expensive part — global 10m embeddings — is now precomputed and free to reuse. You can download the vectors and train a small classifier locally instead of fine-tuning a model with GPU access and storing raw imagery; the new PLANTEUR benchmark is specifically aimed at tropical farming, so it is worth testing the 64-dim embeddings first for a lightweight prototype before committing to a full imagery pipeline.
Discussion angle
If planetary-scale embeddings are precomputed and free, where does the moat move for geospatial SaaS — from model training to local label data, compliance workflows, and last-mile delivery? Pick one Southeast Asian use case such as plantation monitoring, flood mapping, or land-cover compliance and discuss whether the 64-dim or 128-dim embeddings are the better starting point.