According to the research team, the system is the best-performing AI model for Earth observation.
The model excelled at various real-world tasks, including land cover classification, change detection, and multi-sensor analysis.
On average, it outperformed other models by 8% or more.
It instead has an intuitive understanding of geospatial data and our planet.
Thats particularly useful when dealing with an immensely complex system like Earth.
40% off TNW Conference!
The broad dataset covered every region and biome on Earth.
A standout feature of TerraMind is its Thinking-in-Modalities (TiM) tuning.
Similar tochain-of-thought reasoningin language models, TiM lets TerraMind self-generate extra data to improve its performance.
The model is now available open-source on Hugging Face.
Fine-tuned versions will be released in the coming months.
ESA, NASA, and IBM are by no means the only organisations experimenting withAImodels for climate forecasting.
The EU has also experimented with the tech.
Story bySion Geschwindt
Sion is a freelance science and technology reporter, specialising in climate and energy.
