Update app.py

app.py

@@ -6,7 +6,7 @@ from threading import Thread
 import numpy as np
 from PIL import Image
 import subprocess
-import spaces  # Add this import
+import spaces
 
 # Install flash-attention
 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
@@ -106,6 +106,17 @@ def process_vision_query(image, text_input):
     response = vision_processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
     return response
 
+# Combined chat function
+def combined_chat(message, image, history, system_prompt, temperature, max_new_tokens, top_p, top_k):
+    if image is not None:
+        # Process image query
+        response = process_vision_query(image, message)
+        history.append((message, response))
+        return history, None
+    else:
+        # Process text query
+        return stream_text_chat(message, history, system_prompt, temperature, max_new_tokens, top_p, top_k), None
+
 # Custom CSS
 custom_css = """
 body { background-color: #0b0f19; color: #e2e8f0; font-family: 'Arial', sans-serif;}
@@ -114,10 +125,6 @@ body { background-color: #0b0f19; color: #e2e8f0; font-family: 'Arial', sans-ser
 #custom-header h1 .blue { color: #60a5fa;}
 #custom-header h1 .pink { color: #f472b6;}
 #custom-header h2 { font-size: 1.5rem; color: #94a3b8;}
-.suggestions { display: flex; justify-content: center; flex-wrap: wrap; gap: 1rem; margin: 20px 0;}
-.suggestion { background-color: #1e293b; border-radius: 0.5rem; padding: 1rem; display: flex; align-items: center; transition: transform 0.3s ease; width: 200px;}
-.suggestion:hover { transform: translateY(-5px);}
-.suggestion-icon { font-size: 1.5rem; margin-right: 1rem; background-color: #2d3748; padding: 0.5rem; border-radius: 50%;}
 .gradio-container { max-width: 100% !important;}
 #component-0, #component-1, #component-2 { max-width: 100% !important;}
 footer { text-align: center; margin-top: 2rem; color: #64748b;}
@@ -131,28 +138,6 @@ custom_header = """
 </div>
 """
 
-# Custom HTML for suggestions
-custom_suggestions = """
-<div class="suggestions">
-    <div class="suggestion">
-        <span class="suggestion-icon">💬</span>
-        <p>Chat with the Text Model</p>
-    </div>
-    <div class="suggestion">
-        <span class="suggestion-icon">🖼️</span>
-        <p>Analyze Images with Vision Model</p>
-    </div>
-    <div class="suggestion">
-        <span class="suggestion-icon">🤖</span>
-        <p>Get AI-generated responses</p>
-    </div>
-    <div class="suggestion">
-        <span class="suggestion-icon">🔍</span>
-        <p>Explore advanced options</p>
-    </div>
-</div>
-"""
-
 # Gradio interface
 with gr.Blocks(css=custom_css, theme=gr.themes.Base().set(
     body_background_fill="#0b0f19",
@@ -164,34 +149,23 @@ with gr.Blocks(css=custom_css, theme=gr.themes.Base().set(
     block_label_text_color="#94a3b8",
 )) as demo:
     gr.HTML(custom_header)
-    gr.HTML(custom_suggestions)
-
-    with gr.Tab("Text Model (Phi-3.5-mini)"):
-        chatbot = gr.Chatbot(height=400)
-        msg = gr.Textbox(label="Message", placeholder="Type your message here...")
-        with gr.Accordion("Advanced Options", open=False):
-            system_prompt = gr.Textbox(value="You are a helpful assistant", label="System Prompt")
-            temperature = gr.Slider(minimum=0, maximum=1, step=0.1, value=0.8, label="Temperature")
-            max_new_tokens = gr.Slider(minimum=128, maximum=8192, step=1, value=1024, label="Max new tokens")
-            top_p = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=1.0, label="top_p")
-            top_k = gr.Slider(minimum=1, maximum=20, step=1, value=20, label="top_k")
-        
-        submit_btn = gr.Button("Submit", variant="primary")
-        clear_btn = gr.Button("Clear Chat", variant="secondary")
-
-        submit_btn.click(stream_text_chat, [msg, chatbot, system_prompt, temperature, max_new_tokens, top_p, top_k], [chatbot])
-        clear_btn.click(lambda: None, None, chatbot, queue=False)
-
-    with gr.Tab("Vision Model (Phi-3.5-vision)"):
-        with gr.Row():
-            with gr.Column(scale=1):
-                vision_input_img = gr.Image(label="Upload an Image", type="pil")
-                vision_text_input = gr.Textbox(label="Ask a question about the image", placeholder="What do you see in this image?")
-                vision_submit_btn = gr.Button("Analyze Image", variant="primary")
-            with gr.Column(scale=1):
-                vision_output_text = gr.Textbox(label="AI Analysis", lines=10)
-        
-        vision_submit_btn.click(process_vision_query, [vision_input_img, vision_text_input], [vision_output_text])
+
+    chatbot = gr.Chatbot(height=400)
+    msg = gr.Textbox(label="Message", placeholder="Type your message here...")
+    image_input = gr.Image(label="Upload an Image (optional)", type="pil")
+
+    with gr.Accordion("Advanced Options", open=False):
+        system_prompt = gr.Textbox(value="You are a helpful assistant", label="System Prompt")
+        temperature = gr.Slider(minimum=0, maximum=1, step=0.1, value=0.8, label="Temperature")
+        max_new_tokens = gr.Slider(minimum=128, maximum=8192, step=1, value=1024, label="Max new tokens")
+        top_p = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=1.0, label="top_p")
+        top_k = gr.Slider(minimum=1, maximum=20, step=1, value=20, label="top_k")
+    
+    submit_btn = gr.Button("Submit", variant="primary")
+    clear_btn = gr.Button("Clear Chat", variant="secondary")
+
+    submit_btn.click(combined_chat, [msg, image_input, chatbot, system_prompt, temperature, max_new_tokens, top_p, top_k], [chatbot, image_input])
+    clear_btn.click(lambda: ([], None), None, [chatbot, image_input], queue=False)
 
     gr.HTML("<footer>Powered by Phi 3.5 Multimodal AI</footer>")