Update app.py

app.py

@@ -331,11 +331,11 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
                 #download_btn = gr.Button("Download Result.md")
 
             model_choice = gr.Radio(choices=[
-               "Vision-Matters-7B", "MonkeyOCR-1.2B-0709", 
-                "ViGaL-7B", "Visionary-R1-3B", "R1-Onevision-7B"
+                "R1-Onevision-7B", "Vision-Matters-7B",
+                "ViGaL-7B", "MonkeyOCR-1.2B-0709", "Visionary-R1-3B"
             ],
                                     label="Select Model",
-                                    value="Vision-Matters-7B")
+                                    value="R1-Onevision-7B")
 
             gr.Markdown("**Model Info 💻** | [Report Bug](https://huggingface.co/spaces/prithivMLmods/Multimodal-VLMs-5x/discussions)")          
             gr.Markdown("> [MonkeyOCR-1.2B-0709](https://huggingface.co/echo840/MonkeyOCR-1.2B-0709): MonkeyOCR adopts a structure-recognition-relation (SRR) triplet paradigm, which simplifies the multi-tool pipeline of modular approaches while avoiding the inefficiency of using large multimodal models for full-page document processing.")
@@ -343,7 +343,6 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
             gr.Markdown("> [ViGaL 7B](https://huggingface.co/yunfeixie/ViGaL-7B): vigal-7b shows that training a 7b mllm on simple games like snake using reinforcement learning boosts performance on benchmarks like mathvista and mmmu without needing worked solutions or diagrams indicating transferable reasoning skills.")
             gr.Markdown("> [Visionary-R1](https://huggingface.co/maifoundations/Visionary-R1): visionary-r1 is a novel framework for training visual language models (vlms) to perform robust visual reasoning using reinforcement learning (rl). unlike traditional approaches that rely heavily on (sft) or (cot) annotations, visionary-r1 leverages only visual question-answer pairs and rl, making the process more scalable and accessible.")
             gr.Markdown("> [R1-Onevision-7B](https://huggingface.co/Fancy-MLLM/R1-Onevision-7B): r1-onevision model enhances vision-language understanding and reasoning capabilities, making it suitable for various tasks such as visual reasoning and image understanding. with its robust ability to perform multimodal reasoning, r1-onevision emerges as a powerful ai assistant capable of addressing different domains.")
-            
             gr.Markdown(">⚠️note: all the models in space are not guaranteed to perform well in video inference use cases.")  
 
     # Define the submit button actions