Amazon · Compute · Hugging Face
OlmoEarth v1.3: State-of-the-art embeddings for the whole planet
Compiled by KHAO Editorial — aggregated from 1 source. See llms.txt for citation guidance.
★ Tier-1 Source
Models | Tech Report | Data | Code | Embeddings Explorer | Interactive story.
Key facts
- In Peru's Madre de Dios region, illegal gold mining around La Pampa peaked in 2017 and 2018, clearing more than 1,200 hectares of forest in the buffer zone of the Tambopata National Reserve
- Alongside OlmoEarth v1.3, they're sharing precomputed, annual embeddings for the whole planet (10m, 64- and 128-dimensions), plus a new browser-based tool to explore and customize maps based
- OlmoEarth’s 128-dimensional embeddings score highest on 20 of 27 evals and rank highest overall
- The 64-dimensional version, at half the size, still leads on 18 of 27 evals
Summary
Today, they're releasing OlmoEarth v1.3, a state-of-the-art satellite embedding model. Until now, the primary way to use OlmoEarth was fine-tuning in OlmoEarth Studio: training a model for each task and running it over satellite imagery to make a map. With OlmoEarth v1.3, they're excited to give users more ways to explore and engage with planetary artifacts in whatever way suits them best. Foundation models and global embeddings are expensive to produce, and to make sure everyone can access the benefits, they've computed these embeddings for the entire planet and shared them with the community on Source Cooperative. How OlmoEarth v1.3 makes an embedding: a vision transformer reads the satellite data, an Attention Pooler pools it into one vector per patch of land, and a linear layer reduces it to 128 numbers.