Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
This repository contains details of the release of the Prithvi-EO-2.0 foundation model.
| Date | Stars |
|---|---|
| 2026-07-31 | 281 |
| 2026-08-06 | 281 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications
<p align="center">
<img src="https://i.imgur.com/waxVImv.png" alt="Oryx Prithvi-EO-2.0">
</p>
#### [Daniela Szwarcman](https://www.linkedin.com/in/daniela-szwarcman-60b55876/), [Sujit Roy](https://www.linkedin.com/in/sujit-roy01/), [Paolo Fraccaro](https://www.linkedin.com/in/paolo-fraccaro-3b85371b/?originalSubdomain=uk), [Þorsteinn Elí Gíslason](https://www.linkedin.com/in/%C3%BEorsteinn-el%C3%AD-g%C3%ADslason-a6ab951a9), [Benedikt Blumenstiel](https://www.linkedin.com/in/blumenstiel/), [Rinki Ghosal](https://www.linkedin.com/in/rinki-ghosal-5b2a41106/), [Pedro Henrique de Oliveira](https://www.linkedin.com/in/pedro-henrique-conrado-ferreira-de-oliveira-420377220/), [João Lucas de Sousa Almeida](https://www.linkedin.com/in/jo%C3%A3o-lucas-de-sousa-almeida-a08b9255/), [Rocco Sedona](https://www.linkedin.com/in/rocco-sedona-79812749/), [Yanghui Kang](https://www.linkedin.com/in/yanghui-kang-797aa33a/), [Srija Chakraborty](https://www.linkedin.com/in/chakrabortysrija/), [Sizhe Wang](https://scholar.google.com/citations?user=bucEAU0AAAAJ&hl=en), [Ankur Kumar](https://www.linkedin.com/in/ankurk017/), [Myscon Truong](https://www.linkedin.com/in/myscon-truong/), [Denys Godwin](https://www.linkedin.com/in/denys-godwin-43a49188/), [Hyunho Lee](https://scholar.google.com/citations?user=oOwJeyQAAAAJ), [Chia-Yu Hsu](https://www.linkedin.com/in/chiayu-hsu/), [Ata Akbari Asanjan](https://www.linkedin.com/in/ataakbariasanjan/), [Besart Mujeci](https://www.linkedin.com/in/besart/), [Trevor Keenan](https://www.linkedin.com/in/trevor-keenan/), [Paulo Arévolo](https://scholar.google.com/citations?user=AwYBme4AAAAJ&hl=en), [Wenwen Li](https://www.linkedin.com/in/wenwenli/), [Hamed Alemohammad](https://www.linkedin.com/in/hamedalemo/), [Pontus Olofsson](https://www.linkedin.com/in/pontus-olofsson-057701255/), [Christopher Hain](https://www.linkedin.com/in/christopher-hain-5b465917b/), [Robert Kennedy](https://scholar.google.com/citations?user=I-2_GUcAAAAJ&hl=en), [Bianca Zadrozny](https://www.linkedin.com/in/biancazadrozny/), [Gabriele Cavallaro](https://www.linkedin.com/in/dr-gabriele-cavallaro/), [Campbell Watson](https://www.linkedin.com/in/campbell-watson-819101100/), [Manil Maskey](https://www.linkedin.com/in/manilmaskey/), [Rahul Ramachandran](https://www.linkedin.com/in/rramachandran05/), [Juan Bernabe Moreno](https://www.linkedin.com/in/bernabemoreno/)
#### **IBM Research, NASA Marshall Space Flight Center, The University of Alabama in Huntsville, University of Iceland, Jülich Supercomputing Centre, Virginia Tech, Arizona State University, Oregon State University, Clark University, Boston University, University of California, Berkeley, Earth from Space Institute **
[](https://huggingface.co/ibm-nasa-geospatial)
[](https://arxiv.org/abs/2412.02732)
This repository contains code and examples based on the [TerraTorch](https://github.com/IBM/terratorch) library for fine-tuning [Prithvi-EO-2.0](https://huggingface.co/spaces/ibm-nasa-geospatial/Prithvi-EO-2.0-Demo), a more powerful version of the foundation model [Prithvi](https://huggingface.co/ibm-nasa-geospatial/Prithvi-100M) developed by IBM and NASA. Trained on 4.2M global time series samples on the JUWELS HPC system at the Jülich Supercomputing Centre (JSC) using NASA’s Harmonized Landsat and Sentinel data at 30m resolution, it offers significant improvements over its predecessor.
## 📢 Latest Updates
- **December 4, 2024**: Prithvi-EO-2.0 pre-trained models and fine-tuning datasets released on [Hugging Face](https://huggingface.co/ibm-nasa-geospatial).
- **December 5, 2024**: Prithvi-EO-2.0 paper released on [arxiv link](https://arxiv.org/abs/2412.02732). 🔥🔥
## Architecture Overview
Prithvi-EO-2.0 is based on the ViT architecture, pretrained using a mExcerpt of 10,329 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a2af9fb27001aa08, llm:Repository description: 'This repository contains details of the release of the Prithvi-EO-2.0 foundation model.'
matched fp:a2af9fb27001aa08, llm:Repository description: 'This repository contains details of the release of the Prithvi-EO-2.0 foundation model.'