Update app.py

app.py

@@ -1316,13 +1316,17 @@ with gr.Blocks(
 
     gr.Markdown(
         """
-        We introduce **NuExtract 3** — a 4B open-source **MIT License** VLM specialized in document extraction.
-        NuExtract 3 unifies structured extraction — document to JSON — and content extraction — document to Markdown,
-        a.k.a. OCR — into one model.
-
-        NuExtract 3 has been trained via Reinforcement Learning to have extraction-specific reasoning abilities, which can
-        be switched on/off on demand. We find that NuExtract 3 substantially outperforms similar-sized models for both
-        structured extraction and content extraction, making it the new reference model of open-source document extraction.
+        **NuExtract3** is a unified **4B** vision-language reasoning model for document understanding.
+        It combines strong **structured information extraction** with high-quality **image-to-Markdown** conversion, making it suitable for extraction pipelines, OCR, and RAG preprocessing for all types of documents such as scans, receipts, forms, invoices, contracts or tables.
+        
+        ## Overview
+            - **Structured extraction**: input (text/images) + JSON template + instructions --> JSON output
+            - **Markdown conversion**: input (text/images) --> Markdown
+            - **Multimodal inputs**: text, images, or text + images.
+            - **Multilingual** documents.
+            - **Reasoning** and non-reasoning inference modes.
+            - **Template generation** for structured extraction from natural language or input document.
+            
         """,
         elem_classes=["intro-card"],
     )