Update app.py

app.py

@@ -1401,21 +1401,40 @@ with gr.Blocks(
 
     gr.Markdown(
         """
-        **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.
-        
-        **Features**:
-            - **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.
-
-
-        Find the model and inference code here: [NuExtract3](https://huggingface.co/numind/NuExtract3).
-            
-        """,
+<div style="padding: 1rem 0;">
+
+<h1 style="margin-bottom: 0.25rem;">NuExtract3</h1>
+
+<p style="font-size: 1.05rem; line-height: 1.6;">
+<strong>NuExtract3</strong> is a unified <strong>4B vision-language reasoning model</strong>
+for document understanding.
+</p>
+
+<p style="line-height: 1.6;">
+It combines <strong>structured information extraction</strong> with high-quality
+<strong>image-to-Markdown conversion</strong>, making it useful for OCR, RAG preprocessing,
+and extraction pipelines across scans, receipts, forms, invoices, contracts, tables, and more.
+</p>
+
+<h3>Features</h3>
+
+<ul>
+  <li><strong>Structured extraction:</strong> text/images + JSON template + instructions → JSON output</li>
+  <li><strong>Markdown conversion:</strong> text/images → Markdown</li>
+  <li><strong>Multimodal inputs:</strong> text, images, or text + images</li>
+  <li><strong>Multilingual documents</strong></li>
+  <li><strong>Reasoning and non-reasoning inference modes</strong></li>
+  <li><strong>Template generation</strong> from natural language or input documents</li>
+</ul>
+
+<p style="margin-top: 1rem;">
+  <a href="https://huggingface.co/numind/NuExtract3" target="_blank">
+    View model and inference code on Hugging Face →
+  </a>
+</p>
+
+</div>
+    """,
         elem_classes=["intro-card"],
     )