Ai2 releases OlmoEarth satellite map model
SiliconANGLE: Ai2 releases OlmoEarth v1.3 state-of-the-art AI-ready satellite embeddings for the whole planet
Ai2, the Allen Institute for Artificial Intelligence, today released a satellite map embedding model called OlmoEarth v1.3 and precomputed annual embeddings for the whole planet at various resolutions with a browser-based map-making tool. The model compresses a year of satellite observations into a numerical fingerprint for each 10-by-10-meter patch of ground at the smallest resolution, […] The post Ai2 releases OlmoEarth v1.3 state-of-the-art AI-ready satellite embeddings for the whole pl
What happened
The Allen Institute for Artificial Intelligence released OlmoEarth v1.3, a satellite embedding model that compresses a year of satellite observations into numerical fingerprints for 10-by-10-meter patches of ground. The model comes with precomputed annual embeddings for the entire planet at various resolutions and a browser-based tool that allows anyone to analyze satellite data without special resources or sign-in requirements. The model reads radar and light-based imagery from multiple satellite missions.
What each outlet adds
- SiliconANGLE ↗: The new tool requires significantly less computing time and carbon emissions to train than earlier versions and enables practical applications like monitoring illegal mining activity or tracking agricultural expansion.
- Hugging Face ↗: The embeddings ranked highest in performance comparisons against other satellite embedding models across 27 evaluations while also being cheaper and more efficient to produce than previous versions.
Why it matters
The tool allows researchers and organizations to analyze satellite data without needing GPU access, advanced technical knowledge, or the ability to download massive amounts of imagery, making tasks like tracking land changes, detecting specific land uses, or finding similar locations much more accessible. Better global embeddings could improve how AI systems understand and process geographic and environmental data across different regions.
Get smart on it
The Allen Institute for Artificial Intelligence released a satellite map embedding model that compresses a year of satellite observations into numerical fingerprints representing 10-by-10-meter patches of ground, which other software can use to identify crops, forests, mines, and other land cover types. Previously, using this technology required access to large graphics processing units and the ability to download and store imagery, but the new browser-based tool allows anyone to analyze satellite data and track changes over time without special resources or sign-in requirements. The model reads radar and light-based imagery from multiple satellite missions and is more efficient than earlier versions, requiring significantly less computing time and carbon emissions to train. The embeddings enable practical applications like monitoring illegal mining activity or tracking agricultural expansion by comparing satellite data across years without requiring model training for initial analysis.
Relevant courses
Want to go deeper than the news? Explore live, cohort-based AI courses taught by practitioners.
Browse AI courses on MavenRelated news
Mistral AI launches Voxtral 3, a full voice agent stack that listens, thinks, and talks in real time
Mistral AI launched Voxtral 3, a new family of language models built for real-time voice agents. The system combines speech-to-text, text-to-speech, and a voice agent SDK, supports more than 50 languages, and lets agents tap into external tools during live conversations. It can run on your own infrastructure or through EU-hosted APIs to keep data under your control. The article Mistral AI launches Voxtral 3, a full voice agent stack that listens, thinks, and talks in real time appeared
Also reported by MarkTechPost
Beyond tree cover: mapping the world’s forest types
A global map categorizing every 10-meter patch of land on Earth into distinct forest types.
Recursive launches Flint, a laptop coding model it says was largely built by AI
Startup Recursive has released Flint, a free, laptop-sized coding model built largely by its own AI research system, as an early test of AI that improves itself.
Also reported by Wired AI and MarkTechPost
Voxtral 3
Voxtral 3: Mistral's voice agent SDK for real-time speech in 50+ languages — accurate transcription, emotional TTS, deep tool use. EU-hosted or self-deployed.
Hugging Face’s New AI Biology Model Lets You Chat With DNA
The new open-source model speaks the language of DNA and can help researchers understand the effects of any changes on gene activity and health.
Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster
Faster, more accurate speech processing could reduce costs and errors for businesses handling multilingual customer interactions and audio archives.
Also reported by MarkTechPost