Update README.md

README.md

@@ -1,11 +1,12 @@
 # Galeio
 
-Galeio is a company specializing in developing robust foundation models for science and insutrial applications.
+Galeio is a company specializing in developing custom foundation models for science and industrial applications.
 
 ## 🛰️ Our Models
 
 ### The Nereus Family 🌊
-Named after the ancient Greek god of the sea, our Nereus models are specialized in ocean observation:
+Named after the ancient Greek god of the sea, our Nereus models are specialized in ocean observation.
+The models were developed in collaboration with world-leading SAR experts from (Ifremer)[https://en.ifremer.fr/].
 
 #### Current Models
 - **nereus-sar-1**: Our flagship foundation model
@@ -13,18 +14,19 @@ Named after the ancient Greek god of the sea, our Nereus models are specialized
   - State-of-the-art performance on ocean analysis tasks
   - Available architectures:
     - ResNet50 (75.5% TenGeoP accuracy)
-    - ViT-S/16 (78.6% TenGeoP accuracy) (coming soon)
-    - ViT-S/8 (82.1% TenGeoP accuracy)  (coming soon)
-    - ViT-B/8 (83.6% TenGeoP accuracy)  (coming soon)
+    - ViT-S/16 (78.6% TenGeoP accuracy) (on-demand)
+    - ViT-S/8 (82.1% TenGeoP accuracy)  (on-demand)
+    - ViT-B/8 (83.6% TenGeoP accuracy)  (on-demand)
   - Excels at:
-    - Wind speed prediction (RMSE: 1.37)
-    - Wave height estimation (RMSE: 0.63)
+    - Wind speed prediction (RMSE: 1.37 m/s | best model ViT-B/8)  
+    - Wave height estimation (RMSE: 0.63 m | best model ViT-B/8)
     - Geophysical phenomena classification
+    - Many others (benchmarks coming soon)
 
 ### Nereus Family
-The Nereus model family is dedicated to ocean observation and analysis using SAR imagery:
+The Nereus model family is dedicated to ocean observation and analysis using SAR imagery.
 
-- **nereus-sar-1**: Our first foundation model trained on Sentinel-1 Wave Mode data
+- **nereus-sar-1**: Our first foundation model trained on Sentinel-1 Wave Mode data:
   - Specialized for ocean SAR imagery analysis
   - Trained using dynamic dataset pruning for optimal performance
   - Available in multiple architectures (ResNet50, ViT variants)
@@ -39,8 +41,8 @@ Our research focuses on:
 
 ## ✉️ Contact & Support
 
-- 🌐 Website: [galeio.fr](http://galeio.fr)
-- 📧 Email: [email protected]
+- 🌐 Website: [galeio.fr](https://www.galeio.fr)
+- 📧 Email: ([email protected])[mailto:[email protected]]
 - 📍 Location: Paris, France
 - 💼 LinkedIn: [Galeio](https://linkedin.com/company/galeio)