Update app.py

app.py

@@ -304,7 +304,6 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
                 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)")
-        
             model_choice = gr.Radio(
                 choices=["Lumian-VLR-7B-Thinking", "DREX-062225-7B-exp", "olmOCR-7B-0225-preview", "LMM-R1-MGT-PerceReason", "Typhoon-OCR-3B"],
                 label="Select Model",
@@ -314,7 +313,8 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
             gr.Markdown("> Lumian-VLR-7B-Thinking is a high-fidelity vision-language reasoning model built on Qwen2.5-VL-7B-Instruct, designed for fine-grained multimodal understanding, enhancing image captioning, video reasoning, and document comprehension through explicit grounded reasoning. It is trained first via supervised fine-tuning (SFT) on visually-grounded reasoning traces and then further refined using GRPO reinforcement learning to boost reasoning accuracy.")
             gr.Markdown("> LMM-R1-MGT-PerceReason is a vision-language model focused on advanced reasoning using a multimodal tree search approach enabling progressive visual-textual slow thinking, improving complex spatial and logical reasoning without fine-tuning. OLMOCR-7B-0225-preview is a 7B parameter open large model designed for OCR tasks with robust text extraction, especially in complex document layouts. ")
             gr.Markdown("> Typhoon-ocr-3b is a 3B parameter OCR model optimized for efficient and accurate optical character recognition in challenging conditions. DREX-062225-exp is an experimental multimodal model emphasizing strong document reading and extraction capabilities combined with vision-language understanding to support detailed document parsing and reasoning tasks.")
-
+            gr.Markdown("> ⚠️ Note: Models in this space may not perform well on video inference tasks.")
+            
     image_submit.click(
         fn=generate_image,
         inputs=[model_choice, image_query, image_upload, max_new_tokens, temperature, top_p, top_k, repetition_penalty],