Update app.py

app.py

@@ -282,23 +282,18 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
             with gr.Column(elem_classes="canvas-output"):
                 gr.Markdown("## Output")
                 output = gr.Textbox(label="Raw Output Stream", interactive=False, lines=2, show_copy_button=True)
-
                 with gr.Accordion("(Result.md)", open=False):                
                     markdown_output = gr.Markdown(label="(Result.Md)")
-                #download_btn = gr.Button("Download Result.md"
-                
+        
             model_choice = gr.Radio(
                 choices=["Lumian-VLR-7B-Thinking", "DREX-062225-7B-exp", "olmOCR-7B-0225-preview", "Typhoon-OCR-3B"],
                 label="Select Model",
-                value="DREX-062225-7B-exp"
+                value="Lumian-VLR-7B-Thinking"
             )
 
-            gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Doc-VLMs/discussions)")          
-            gr.Markdown("> [DREX-062225-7B-exp](https://huggingface.co/prithivMLmods/DREX-062225-exp): the drex-062225-exp (document retrieval and extraction expert) model is a specialized fine-tuned version of docscopeocr-7b-050425-exp, optimized for document retrieval, content extraction, and analysis recognition. built on top of the qwen2.5-vl architecture.")
-            gr.Markdown("> [Megalodon-OCR-Sync-0713](https://huggingface.co/prithivMLmods/Megalodon-OCR-Sync-0713): megalodon-ocr-sync-0713 model is a fine-tuned version of qwen2.5-vl-3b-instruct, optimized for document retrieval, content extraction, and analysis recognition. built on top of the qwen2.5-vl architecture, this model enhances document comprehension capabilities.")
-            gr.Markdown("> [Typhoon-OCR-3B](https://huggingface.co/scb10x/typhoon-ocr-3b): a bilingual document parsing model built specifically for real-world documents in thai and english, inspired by models like olmocr, based on qwen2.5-vl-instruction. this model is intended to be used with a specific prompt only.")
-            gr.Markdown("> [olmOCR-7B-0225](https://huggingface.co/allenai/olmOCR-7B-0225-preview): the olmocr-7b-0225-preview model is based on qwen2-vl-7b, optimized for document-level optical character recognition (ocr), long-context vision-language understanding, and accurate image-to-text conversion with mathematical latex formatting. designed with a focus on high-fidelity visual-textual comprehension.")
-            gr.Markdown(">⚠️note: all the models in space are not guaranteed to perform well in video inference use cases.")  
+            gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Multimodal-VLM-Thinking/discussions)")          
+            gr.Markdown("> Lumian-VLR-7B-Thinking is a high-fidelity vision-language reasoning model built on Qwen2.5-VL-7B-Instruct, excelling at fine-grained multimodal tasks such as image captioning, sampled video reasoning, and document comprehension through explicit grounded reasoning and advanced reinforcement learning. olmOCR-7B-0225-preview, developed by AllenAI, is a Qwen2-VL-7B-Instruct derivative optimized specifically for robust document OCR, efficiently processing large volumes of document images with specialized prompting and high scalability.")
+            gr.Markdown(">  Typhoon-OCR-3B targets bilingual (Thai and English) document parsing, providing reliable OCR and text extraction for real-world documents, emphasizing usability in diverse and complex layouts. DREX-062225-exp is a document retrieval and extraction expert model, fine-tuned from docscopeOCR-7B, focusing on superior document analysis, structured data extraction, and maintaining advanced OCR capabilities including LaTeX and multilingual support. Together, these models represent the state-of-the-art in multimodal document understanding, OCR, and vision-language reasoning for a wide range of real-world and research applications.")
     
     image_submit.click(
         fn=generate_image,
@@ -310,11 +305,6 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
         inputs=[model_choice, video_query, video_upload, max_new_tokens, temperature, top_p, top_k, repetition_penalty],
         outputs=[output, markdown_output]
     )
-#    download_btn.click(
-#        fn=save_to_md,
-#        inputs=output,
-#        outputs=None
-#    )
 
 if __name__ == "__main__":
     demo.queue(max_size=30).launch(share=True, mcp_server=True, ssr_mode=False, show_error=True)