Update app.py

app.py

@@ -41,8 +41,9 @@ with gr.Blocks(title="MedSigLIP Classifier") as demo:
         gr.Markdown(
             """
             # MedSigLIP Zero-Shot Classification
-            MedSigLIP is a medical version of SigLIP, trained to map medical images and text into a shared embedding space. It features 400M vision + 400M text encoders, supports 448×448 images and 64 text tokens, and is trained on diverse de-identified medical image–text pairs plus natural images. Best suited for medical image interpretation without text generation — e.g., data-efficient classification, zero-shot classification, and semantic image retrieval. For applications requiring report or text generation, MedGemma is recommended.
-        
+            MedSigLIP is a medical adaptation of SigLIP that embeds medical images and text into a shared space. It uses 400M vision + 400M text encoders, supports 448×448 images and 64 text tokens, and is trained on broad de-identified medical image–text pairs plus natural images.
+            Best for medical image tasks like data-efficient classification, zero-shot classification, and semantic retrieval.
+            
             ## Links
             * Model Page: https://huggingface.co/google/medsiglip-448
             * Model Documentation: https://developers.google.com/health-ai-developer-foundations/medsiglip