Update app.py

app.py

@@ -392,127 +392,66 @@ class StreamProcessor:
             # Handle thinking tags
             if not in_thought and '<think>' in text:
                 in_thought = True
-                history.append({'role': 'assthought = True
-                history.append({'role': 'assistant', 'contentistant', 'content': '', 'metadata': {'title': '💭 Thought'}})
-                assistant_message_started': '', 'metadata': {'title': '💭 Thought' = True
-                after = text.split('<think>', 1)[1]
-                thought_buf += after}})
+                history.append({'role': 'assistant', 'content': '', 'metadata': {'title': '💭 Thought'}})
                 assistant_message_started = True
-                after = text.split('<think>', 1)[1
-                
-                if '</think>' in thought_buf:
-                    before,]
+                after = text.split('<think>', 1)[1]
                 thought_buf += after
                 
                 if '</think>' in thought_buf:
-                    after2 = thought_buf.split('</think>', 1)
                     before, after2 = thought_buf.split('</think>', 1)
- history[-1]['content']                    history[-1]['content'] = before.strip()
-                    in_thought = False
-                    answer_buf = = before.strip()
+                    history[-1]['content'] = before.strip()
                     in_thought = False
                     answer_buf = after2
-                    history.append({'role': ' after2
-                    history.append({'roleassistant', 'content': answer_buf})
-                else:
-                    history[-1]['': 'assistant', 'content': answer_buf})
+                    history.append({'role': 'assistant', 'content': answer_buf})
                 else:
-                   content'] = thought_buf
-                yield history, ""
-                continue
-            
-            if in_thought:
-                thought_b history[-1]['content'] = thought_buf
+                    history[-1]['content'] = thought_buf
                 yield history, ""
                 continue
             
             if in_thought:
                 thought_buf += text
                 if '</think>' in thought_buf:
-                    beforeuf += text
-                if '</think>' in thought_buf:
-                   , after2 = thought_buf.split('</think>', 1)
- before, after2 = thought_buf.split('</think>', 1)
+                    before, after2 = thought_buf.split('</think>', 1)
                     history[-1]['content'] = before.strip()
-                    in_thought =                    history[-1]['content'] = before.strip()
-                    in_thought False
-                    answer_buf = after2
-                    history.append({'role': = False
+                    in_thought = False
                     answer_buf = after2
-                    history 'assistant', 'content': answer_buf})
-                else:
                     history.append({'role': 'assistant', 'content': answer_buf})
                 else:
-                    history[-1]['content[-1]['content'] = thought_buf
-                yield history, ""
-'] = thought_buf
+                    history[-1]['content'] = thought_buf
                 yield history, ""
                 continue
             
             # Stream answer
             if not assistant_message_started:
-                history.append({'role                continue
-            
-            # Stream answer
-            if not assistant_message_started:
-               ': 'assistant', 'content': ''})
-                assistant_message_start history.append({'role': 'assistant', 'content': ''})
+                history.append({'role': 'assistant', 'content': ''})
                 assistant_message_started = True
             
-            answered = True
-            
             answer_buf += text
             history[-1]['content'] = answer_buf.strip()
-            yield history, "_buf += text
-            history[-1]['content'] = answer_buf.strip()
             yield history, ""
 
-# Main chat function
-def chat_response(
-    user_msg: str,
-    chat"
-
 # Main chat function
 def chat_response(
     user_msg: str,
     chat_history: List[Dict],
     system_prompt: str,
-_history: List[Dict],
-    system_prompt: str,
-    enable_search    enable_search: bool,
-    max_results: int,
-    max_ch: bool,
+    enable_search: bool,
     max_results: int,
     max_chars: int,
-ars: int,
     model_name: str,
     max_tokens: int,
     temperature: float,
-    top    model_name: str,
-    max_tokens: int,
-   _k: int,
-    top_p temperature: float,
     top_k: int,
     top_p: float,
-    repeat: float,
     repeat_penalty: float,
     search_timeout: float
-)_penalty: float,
-    search_timeout: float
 ) -> Generator[Tuple[List[Dict], str], None, None]:
-    """Generate -> Generator[Tuple[List[Dict], str], None, None]:
     """Generate streaming chat responses with search integration"""
     
-    cancel streaming chat responses with search integration"""
-    
     cancel_event.clear()
     history = list(chat_history or [])
     history.append({'role': 'user', 'content': user_msg})
     
-    #_event.clear()
-    history = list(chat_history or [])
-    history.append({'role': 'user', 'content': user_msg})
-    
     # Perform search if enabled
     search_results: List[SearchResult] = []
     search_debug = "Web search disabled."
@@ -523,44 +462,18 @@ ars: int,
             search_results = SearchManager.search(
                 user_msg, 
                 int(max_results), 
-                Perform search if enabled
-    search_results: List[SearchResult] = []
-    search_debug = "Web search disabled."
-    
-    if enable_search:
-        search_debug = "🔍 Searching across multiple engines..."
-        try:
-            search_results = SearchManager.search(
-                user_msg, 
-                int(max int(max_chars), 
-                float(search_timeout)
-            )
-            
-            if search_results:
-_results), 
                 int(max_chars), 
                 float(search_timeout)
             )
             
             if search_results:
-                search_debug = f"✅ Search completed - Found {len(search_results)} results\n\n" + "\n".join                search_debug = f"✅ Search completed - Found {len(search_results)} results\n\n" + "\n".join(
-                    f"- {r.format(int(max_chars))}" for r in(
+                search_debug = f"✅ Search completed - Found {len(search_results)} results\n\n" + "\n".join(
                     f"- {r.format(int(max_chars))}" for r in search_results
                 )
             else:
-                search_debug = "❌ No search search_results
-                )
-            else:
-                search_debug = "❌ No search results found. results found. Check internet connection or try again."
+                search_debug = "❌ No search results found. Check internet connection or try again."
         except Exception as e:
-            search_debug = f"❌ Check internet connection or try again."
-        except Exception as e:
-            search Search failed: {str(e)}"
-            logger.error(f"Search error: {e}")
-    
-    try:
-        # Build prompt
-        if enable_debug = f"❌ Search failed: {str(e)}"
+            search_debug = f"❌ Search failed: {str(e)}"
             logger.error(f"Search error: {e}")
     
     try:
@@ -568,53 +481,28 @@ _results),
         if enable_search and search_results:
             enriched_prompt = PromptBuilder.build_search_context(
                 search_results, 
-               _search and search_results:
-            enriched_prompt = PromptBuilder.build_search_context(
-                search_results, 
-                system_prompt, system_prompt, 
+                system_prompt, 
                 user_msg
             )
         else:
-            enriched_prompt = system_prompt 
-                user_msg
-            )
-        else:
-            enriched_prompt = system.strip()
-        
-        # Load model
-        pipe = ModelManager.load_pipeline(model_name)
-        
-        # Format prompt
-        prompt = PromptBuilder.format_conversation_prompt.strip()
+            enriched_prompt = system_prompt.strip()
         
         # Load model
         pipe = ModelManager.load_pipeline(model_name)
         
         # Format prompt
-       (history, enriched_prompt, pipe.tokenizer)
-        prompt_debug = f"\n\n--- Prompt prompt = PromptBuilder.format_conversation(history, enriched_prompt, pipe.tokenizer)
-        prompt_debug = f"\n\n--- Prompt Preview ---\n```\n{prompt[:500]}...\n``` Preview ---\n```\n{prompt[:500]}...\n" if len(prompt) > 500 else f"\n```" if len(prompt) > 500 else f"\n\n--- Prompt Preview\n--- Prompt Preview ---\n```\n{prompt}\n```"
+        prompt = PromptBuilder.format_conversation(history, enriched_prompt, pipe.tokenizer)
+        prompt_debug = f"\n\n--- Prompt Preview ---\n```\n{prompt[:500]}...\n```" if len(prompt) > 500 else f"\n\n--- Prompt Preview ---\n```\n{prompt}\n```"
         
         # Configure generation
         config = GenerationConfig(
-            max_tokens ---\n```\n{prompt}\n```"
-        
-        # Configure generation
-        config = GenerationConfig(
-            max_tokens=max_t=max_tokens,
-            temperature=temperatureokens,
+            max_tokens=max_tokens,
             temperature=temperature,
             top_k=top_k,
             top_p=top_p,
             repetition_penalty=repeat_penalty
         )
         
-        # Setup,
-            top_k=top_k,
-            top_p=top_p,
-            repetition_penalty=repeat_penalty
-        )
-        
         # Setup streamer
         streamer = TextIteratorStreamer(
             pipe.tokenizer,
@@ -622,212 +510,105 @@ _results),
             skip_special_tokens=True
         )
         
-        # Start streamer
-        streamer = TextIteratorStreamer(
-            pipe.tokenizer,
-            skip_prompt=True,
-            skip_special_tokens=True
-        )
-        
-        # Start generation in generation in background thread
+        # Start generation in background thread
         gen_kwargs = config.to_dict()
-        gen_kwargs['stream background thread
-        gen_kwargs = config.to_dict()
-        gener'] = streamer
-        gen_kwargs['return_full_text_kwargs['streamer'] = streamer
+        gen_kwargs['streamer'] = streamer
         gen_kwargs['return_full_text'] = False
         
-       '] = False
-        
         gen_thread = threading.Thread(
-            target= gen_thread = threading.Thread(
             target=pipe,
             args=(prompt,),
             kwargs=gen_kwargs
         )
         gen_thread.start()
         
-        # Yield initialpipe,
-            args=(prompt,),
-            kwargs=gen_kwargs
-        )
-        gen_thread.start()
-        
-        # Yield initial state state
-        yield history, search_debug
-        
-        # Process stream
-        for
+        # Yield initial state
         yield history, search_debug
         
         # Process stream
         for history_update, debug_update in StreamProcessor.process_stream(streamer, history):
-            yield history_update history_update, debug_update in StreamProcessor.process_stream(streamer, history):
             yield history_update, debug_update
         
         # Wait for completion
-       , debug_update
-        
-        # Wait for completion
         gen_thread.join(timeout=5.0)
         yield history, search_debug + prompt_debug
         
-    gen_thread.join(timeout=5.0)
-        yield history, search_debug + prompt_debug
-        
     except GeneratorExit:
-        logger.info("Generation except GeneratorExit:
         logger.info("Generation cancelled by user")
         return
-    except Exception as e:
-        logger.error(f" cancelled by user")
-        return
     except Exception as e:
         logger.error(f"Generation error: {e}")
-        history.append({'role': 'assistant', 'content': f"Error: {strGeneration error: {e}")
-        history.append({'role': 'assistant', 'content': f"(e)}"})
+        history.append({'role': 'assistant', 'content': f"Error: {str(e)}"})
         yield history, search_debug
     finally:
         gc.collect()
 
 # Utility functions
 def get_model_size(model_name: str) -> float:
-    """Error: {str(e)}"})
-        yield history, search_debug
-    finally:
-        gc.collect()
-
-# Utility functions
-def get_model_size(model_name: str) ->Get model size in billions of parameters"""
-    return MODELS.get(model_name, {}).get("params float:
     """Get model size in billions of parameters"""
-    return MODELS.get(model_name, {}).get("params_b", 4_b", 4.0)
-
-def get_duration_estimate(
-    model_name: str,
-    enable_search: bool,
-    max_t.0)
+    return MODELS.get(model_name, {}).get("params_b", 4.0)
 
 def get_duration_estimate(
     model_name: str,
     enable_search: bool,
     max_tokens: int,
     search_timeout: float
-) -> float:
-    """Calculateokens: int,
-    search_timeout: float
 ) -> float:
     """Calculate estimated GPU duration"""
     model_size = get_model_size(model_name)
- estimated GPU duration"""
-    model_size = get_model_size(model_name)
     use_aot = model_size >= 2
     
-    base    use_aot = model_size >= 2
-    
-   _duration = 20 if not use_aot else 40
-    token_duration = max_tokens *  base_duration = 20 if not use_aot else 40
+    base_duration = 20 if not use_aot else 40
     token_duration = max_tokens * 0.005
     search_duration = 10 if enable_search else 0
-    a0.005
-    search_duration = 10 if enable_search else 0
     aot_compilation = 20 if use_aot else 0
     
-    return base_dot_compilation = 20 if use_aot else 0
-    
     return base_duration + token_duration + search_duration + aot_compilation
 
-def update_duration_estimate(
-    modeluration + token_duration + search_duration + aot_compilation
-
 def update_duration_estimate(
     model_name: str,
-    enable_name: str,
     enable_search: bool,
     max_results: int,
-    max_chars: int_search: bool,
-    max_results: int,
     max_chars: int,
     max_tokens: int,
     search_timeout: float
 ) -> str:
     """Format duration estimate for display"""
     try:
-        duration = get_duration_estimate(model_name, enable_search, max_tokens,
-    max_tokens: int,
-    search_timeout: float
-) -> str:
-    """Format duration estimate for display"""
-    try:
-        duration = get_duration_estimate(model_name, enable_search, max, search_timeout)
+        duration = get_duration_estimate(model_name, enable_search, max_tokens, search_timeout)
         model_size = get_model_size(model_name)
         
-        return f"""⏱_tokens, search_timeout)
-        model_size = get_model_size(model_name)
-        
-        return f️ **Estimated GPU Time: {duration:.1f} seconds**
+        return f"""⏱️ **Estimated GPU Time: {duration:.1f} seconds**
 
 📊 **Model Size:** {model_size:.1f}B parameters
-🔍"""⏱️ **Estimated GPU Time: {duration:.1f} seconds**
-
-📊 **Model Size:** {model_size:.1f}B parameters
-🔍 ** **Web Search:** {'Enabled (Multi-Engine)' if enable_search else 'Disabled'}"""
-    except ExceptionWeb Search:** {'Enabled (Multi-Engine)' if enable_search else 'Disabled'}"""
+🔍 **Web Search:** {'Enabled (Multi-Engine)' if enable_search else 'Disabled'}"""
     except Exception as e:
-        logger.error(f"Error calculating estimate: {e}")
-        return f"⚠️ Error as e:
         logger.error(f"Error calculating estimate: {e}")
         return f"⚠️ Error calculating estimate: {e}"
 
 def update_default_prompt(enable_search: bool) -> str:
-    """Generate calculating estimate: {e}"
-
-def update_default_prompt(enable_search: bool) -> str:
-    """ default system prompt"""
-    return "You are aGenerate default system prompt"""
+    """Generate default system prompt"""
     return "You are a helpful assistant."
 
 # ------------------------------
-# Gradio UI (unchanged)
-# ------------------------------
-with gr helpful assistant."
-
-# ------------------------------
-# Gradio UI (unchanged)
+# Gradio UI
 # ------------------------------
 with gr.Blocks(
-.Blocks(
     title="LLM Inference",
-    theme=gr.themes.Soft(
-        primary_hue="blue",
-        secondary_hue="blue",
-        neutral_hue="slate",
-        radius    title="LLM Inference",
     theme=gr.themes.Soft(
         primary_hue="blue",
         secondary_hue="blue",
         neutral_hue="slate",
         radius_size="lg",
-        font=[gr_size="lg",
         font=[gr.themes.GoogleFont("Syne"), "Arial", "sans-serif"]
     ),
     css="""
-       .themes.GoogleFont("Syne"), "Arial", "sans-serif"]
-    ),
-    css="""
- .duration-estimate { background: linear-gradient(135deg, #667eea15         .duration-estimate { background: linear-gradient(135deg, #667eea15 0%, #0%, #764ba215 100%); border-left: 4764ba215 px solid #667eea; padding: 12px; border-radius:100%); border-left: 4px solid #667eea; padding: 12px; border-radius: 8px; margin: 16px 0; }
-        . 8px; margin: 16px 0; }
-        .chatbot { border-radius: 12px; box-shadow: 0 4px 6px -chatbot { border-radius: 12px; box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1); }
-        button.primary {1px rgba(0, 0, 0, 0.1); }
-        button.primary { font font-weight: 600; }
-        .gradio-accordion { margin-bottom: 12px; }
-    """
--weight: 600; }
+        .duration-estimate { background: linear-gradient(135deg, #667eea15 0%, #764ba215 100%); border-left: 4px solid #667eea; padding: 12px; border-radius: 8px; margin: 16px 0; }
+        .chatbot { border-radius: 12px; box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1); }
+        button.primary { font-weight: 600; }
         .gradio-accordion { margin-bottom: 12px; }
     """
 ) as demo:
-    # Header
-    gr.Markdown("") as demo:
     # Header
     gr.Markdown("""
     # 🧠 LLM Inference with Multi-Engine Search
@@ -835,138 +616,67 @@ with gr.Blocks(
     
     with gr.Row():
         # Left Panel - Configuration
-        with gr.Column(scale"
-    # 🧠 LLM Inference with Multi-Engine Search
-    """)
-    
-    with gr.Row():
-        # Left Panel - Configuration
-        with gr.Column(=3):
-            # Core Settings (Always Visible)
-            with gr.Group():
-                gr.Markdown("scale=3):
+        with gr.Column(scale=3):
             # Core Settings (Always Visible)
             with gr.Group():
-                gr.Markdown### ⚙️ Core Settings")
-                model_dd = gr.D("### ⚙️ Core Settings")
-                model_dd = grropdown(
-                    label="🤖 Model",
-                    choices=list(MODELS.keys()),
-                    value="Qwen3.Dropdown(
+                gr.Markdown("### ⚙️ Core Settings")
+                model_dd = gr.Dropdown(
                     label="🤖 Model",
                     choices=list(MODELS.keys()),
                     value="Qwen3-1.7B",
                     info="Select the language model to use"
                 )
-                search_chk = gr.Check-1.7B",
-                    info="Select the language model to use"
-                )
                 search_chk = gr.Checkbox(
                     label="🔍 Enable Web Search",
                     value=False,
-                    info="Search across Googlebox(
-                    label="🔍 Enable Web Search",
-                    value=False,
-                    info="Search across Google,, DuckDuckGo, and Bing (no API required)"
-                )
-                sys_prompt = gr.Textbox(label="📝 System Prompt", lines=3, value=update_default_prompt(False), placeholder="Define the assistant's behavior and personality...")
-            
-            # DuckDuckGo, and Bing (no API required)"
+                    info="Search across Google, DuckDuckGo, and Bing (no API required)"
                 )
                 sys_prompt = gr.Textbox(label="📝 System Prompt", lines=3, value=update_default_prompt(False), placeholder="Define the assistant's behavior and personality...")
             
- Duration Estimate
+            # Duration Estimate
             duration_display = gr.Markdown(
-                value=update_duration_estimate("Qwen3-1.            # Duration Estimate
-            duration_display = gr.Markdown(
-                value=update_duration_estimate("Qwen3-1.7B", False7B", False, 4, 50, 1024, 5.0),
-                elem_classes="duration-est, 4, 50, 1024, 5.0),
+                value=update_duration_estimate("Qwen3-1.7B", False, 4, 50, 1024, 5.0),
                 elem_classes="duration-estimate"
-           imate"
             )
             
             # Advanced Settings (Collapsible)
             with gr.Accordion("🎛️ Advanced Generation Parameters", open=False):
- )
-            
-            # Advanced Settings (Collapsible)
-            with gr.Accordion("🎛️ Advanced Generation                max_tok = gr.Slider(
-                    64, 16384, value=1024, step=32,
-                    label="Max Tok Parameters", open=False):
                 max_tok = gr.Slider(
                     64, 16384, value=1024, step=32,
                     label="Max Tokens",
-                    infoens",
                     info="Maximum length of generated response"
                 )
-                temp = gr.Slider(
-                    0.1, ="Maximum length of generated response"
-                )
                 temp = gr.Slider(
                     0.1, 2.0, value=0.7, step=0.1,
-                    label="Temperature2.0, value=0.7, step=0.1,
                     label="Temperature",
                     info="Higher = more creative, Lower = more focused"
                 )
-",
-                    info="Higher = more creative, Lower = more focused"
-                )
                 with gr.Row():
                     k = gr.Slider(
-                        1, 100, value=40, step=1                with gr.Row():
-                    k = gr.Slider(
-                        1, 100, value=40,
-                        label="Top-K",
-                        info="Number of top tokens to consider"
-                    )
-                    p = gr.Slider, step=1,
+                        1, 100, value=40, step=1,
                         label="Top-K",
                         info="Number of top tokens to consider"
                     )
                     p = gr.Slider(
-                        (
                         0.1, 1.0, value=0.9, step=0.05,
                         label="Top-P",
                         info="Nucleus sampling threshold"
                     )
-                rp = gr.S0.1, 1.0, value=0.9, step=0.05,
-                        label="Top-P",
-                        info="Nucleus sampling threshold"
-                    )
                 rp = gr.Slider(
-                    lider(
-                    1.0,1.0, 2.0, value=1.2, step=0.1,
-                    label="Repetition Penalty",
-                    info="Penalize 2.0, value=1.2, step=0.1,
+                    1.0, 2.0, value=1.2, step=0.1,
                     label="Repetition Penalty",
-                    info repeated tokens"
-                )
-            
-            # Web Search Settings (Collapsible)
-            with gr.Accordion("🌐 Web Search Settings", open=False,="Penalize repeated tokens"
+                    info="Penalize repeated tokens"
                 )
             
             # Web Search Settings (Collapsible)
             with gr.Accordion("🌐 Web Search Settings", open=False, visible=False) as search_settings:
-                mr = gr.Number(
-                    value=4, precision=0,
-                    label="Max Results",
-                    info="Number of visible=False) as search_settings:
                 mr = gr.Number(
                     value=4, precision=0,
                     label="Max Results",
                     info="Number of search results to retrieve"
                 )
                 mc = gr.Number(
-                    value=50, precision search results to retrieve"
-                )
-                mc = gr.Number(
-                    value=50, precision==0,
-                    label="Max Chars/Result",
-                    info="Character limit per search result"
-                )
-                st = gr.Slider(
-                    minimum=0.0, maximum=30.0, step=0.50,
+                    value=50, precision=0,
                     label="Max Chars/Result",
                     info="Character limit per search result"
                 )
@@ -976,25 +686,10 @@ with gr.Blocks(
                     info="Maximum time to wait for search results"
                 )
                 gr.Markdown("""
-               , value=5.0,
-                    label="Search Timeout (s)",
-                    info="Maximum time to wait for search results"
-                )
-                gr.Markdown("" ⚠️ **Search Engines:**
-                - Google (primary)
-                - DuckDuckGo (fallback)
-                - Bing (fallback)
-                
-"
                 ⚠️ **Search Engines:**
                 - Google (primary)
                 - DuckDuckGo (fallback)
-                - Bing (                SafeSearch is **OFF** for comprehensive results.
-                """)
-            
-            # Actions
-            with gr.Row():
-                clr = gr.Button("🗑️ Clearfallback)
+                - Bing (fallback)
                 
                 SafeSearch is **OFF** for comprehensive results.
                 """)
@@ -1005,145 +700,72 @@ with gr.Blocks(
         
         # Right Panel - Chat Interface
         with gr.Column(scale=7):
-            chat = Chat", variant="secondary", scale=1)
-        
-        # Right Panel - Chat Interface
-        with gr.Column(scale=7):
- gr.Chatbot(
-                type="messages",
-                height=600,
-                label="💬 Conversation",
-                show            chat = gr.Chatbot(
+            chat = gr.Chatbot(
                 type="messages",
                 height=600,
                 label="💬 Conversation",
                 show_copy_button=True,
-_copy_button=True,
                 avatar_images=(
-                    "data:image                avatar_images=(
-                    "data:image/svg+xml,%3/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40'%3E%3Crect width='40' height='40' rx='20' fill='%23f093fb'/%3E%3Ctext x'%3E%3Crect width='40' height='40' rx='20' fill='%23f093fb'/%3E%3Ctext x='20' y='28' text-anchor='middle' font-size='20' fill='='20' y='28' text-anchor='middle' font-size='20' fill='white' font-family='Arial'%3E👤%3C/text%3E%3C/swhite' font-family='Arial'%3E👤%3C/text%3E%3C/svg%3E",
-                    "data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40'vg%3E",
-                    "data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40'%3E%3Crect width='40' height='40' rx='20' fill='%23667eea'/%3E%3Ctext x='20' y='28' text-anchor='middle' font-size='20' fill='white' font-family='Arial'%%3E%3Crect width='40' height='40' rx='20' fill='%23667eea'/%3E%3Ctext x='20' y='28' text-anchor='middle' font-size='20' fill='white' font-family='3E🤖%3C/text%3E%3C/svg%3E"
-                ),
-Arial'%3E🤖%3C/text%3E%3C/svg%3E"
+                    "data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40'%3E%3Crect width='40' height='40' rx='20' fill='%23f093fb'/%3E%3Ctext x='20' y='28' text-anchor='middle' font-size='20' fill='white' font-family='Arial'%3E👤%3C/text%3E%3C/svg%3E",
+                    "data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='40' height='40'%3E%3Crect width='40' height='40' rx='20' fill='%23667eea'/%3E%3Ctext x='20' y='28' text-anchor='middle' font-size='20' fill='white' font-family='Arial'%3E🤖%3C/text%3E%3C/svg%3E"
                 ),
                 bubble_full_width=False,
-                bubble_full_width=False,
                 render_markdown=True,
                 sanitize_html=False
             )
             
             # Input Area
             with gr.Row():
-                               render_markdown=True,
-                sanitize_html=False
-            )
-            
-            # Input Area
-            with gr txt = gr.Textbox(
-                    placeholder="💭 Type your message here... (Press Enter to.Row():
                 txt = gr.Textbox(
                     placeholder="💭 Type your message here... (Press Enter to send)",
-                    send)",
                     scale=9,
                     container=False,
                     show_label=False,
                     lines=1,
-                    max scale=9,
-                    container=False,
-                    show_label=False,
-                    lines=1,
-                    max_lines=5_lines=5
-                )
-                with gr.Column(scale=1, min_width=120):
-                    submit_btn
+                    max_lines=5
                 )
                 with gr.Column(scale=1, min_width=120):
                     submit_btn = gr.Button("📤 Send", variant="primary", size="lg")
-                    cancel_btn = gr.Button("⏹️ Stop", variant=" = gr.Button("📤 Send", variant="primary", size="lg")
-                    cancel_btn = gr.Button("⏹️stop", visible=False, size="lg")
-            
-            # Example Prompts
-            gr.Examples(
-                Stop", variant="stop", visible=False, size="lg")
+                    cancel_btn = gr.Button("⏹️ Stop", variant="stop", visible=False, size="lg")
             
             # Example Prompts
             gr.Examples(
                 examples=[
-                    ["Explain examples=[
                     ["Explain quantum computing in simple terms"],
                     ["Write a Python function to calculate fibonacci numbers"],
-                    ["What are the latest developments in AI quantum computing in simple terms"],
-                    ["Write a Python function to calculate fibonacci numbers"],
-                    ["? (Enable web search)"],
-                    ["Tell me a creative story about a time traveler"],
-                    ["Help me debug this code: def add(a,bWhat are the latest developments in AI? (Enable web search)"],
+                    ["What are the latest developments in AI? (Enable web search)"],
                     ["Tell me a creative story about a time traveler"],
-                    ["Help me debug this code: def add(a,b):): return a+b+1"]
+                    ["Help me debug this code: def add(a,b): return a+b+1"]
                 ],
                 inputs=txt,
-                label="💡 Example Prom return a+b+1"]
-                ],
-                inputs=txt,
-                label="pts"
-            )
-            
-            # Debug/Status Info (Collapsible)
-            with gr.Accordion("🔍 Debug Info", open=False):
-                dbg = gr💡 Example Prompts"
+                label="💡 Example Prompts"
             )
             
             # Debug/Status Info (Collapsible)
             with gr.Accordion("🔍 Debug Info", open=False):
                 dbg = gr.Markdown()
     
-    # Footer.Markdown()
-    
     # Footer
     gr.Markdown("""
     ---
     💡 **Tips:** 
-    - Use **Advanced Parameters** to fine-tune creativity and response
-    gr.Markdown("""
-    ---
-    💡 **Tips:** 
     - Use **Advanced Parameters** to fine-tune creativity and response length
- length
     - Enable **Web Search** for real-time information (uses multiple search engines)
-    - SafeSearch is **OFF** for comprehensive results    - Enable **Web Search** for real-time information (uses multiple search engines)
     - SafeSearch is **OFF** for comprehensive results
-    - Try different **
-    - Try different **models** for various tasks (reasonmodels** for various tasks (reasoning, coding, general chat)
+    - Try different **models** for various tasks (reasoning, coding, general chat)
     - Click the **Copy** button on responses to save them to your clipboard
-ing, coding, general chat)
-    - Click the **Copy** button on responses to save them to your clipboard
-    """, elem_classes="footer    """, elem_classes="footer")
-
-    # --- Event Listeners ---
-
-    # Group all inputs")
+    """, elem_classes="footer")
 
     # --- Event Listeners ---
 
-    # Group all for cleaner event handling
+    # Group all inputs for cleaner event handling
     chat_inputs = [txt, chat, sys_prompt, search_chk, mr, mc, model_dd, max_tok, temp, k, p, rp, st]
-    # Group inputs for cleaner event handling
-    chat_inputs = [txt, chat, sys_prompt, search_chk, mr, mc, model_dd, max_tok, temp, k, p, rp, st all UI components that can be updated.
-    ui_components = [chat, dbg, txt, submit_btn, cancel_btn]
     # Group all UI components that can be updated.
     ui_components = [chat, dbg, txt, submit_btn, cancel_btn]
 
-    def submit]
-
     def submit_and_manage_ui(user_msg, chat_history, *args):
         """
-        Orchestrator_and_manage_ui(user_msg, chat_history, *args):
-        """
-        Orchestrator function that manages UI function that manages UI state and calls the backend chat function.
-        """
-        if not user_msg.strip():
-            yield {}
- state and calls the backend chat function.
+        Orchestrator function that manages UI state and calls the backend chat function.
         """
         if not user_msg.strip():
             yield {}
@@ -1156,175 +778,86 @@ ing, coding, general chat)
             cancel_btn: gr.update(visible=True),
         }
 
-        cancelled =            return
-
-        # Update UI to "generating" state
-        yield {
-            txt: gr.update(value="", interactive=False),
-            submit_btn: gr.update(interactive=False),
-            cancel_btn: gr.update(visible=True),
-        }
-
         cancelled = False
         try:
             backend_args = [user_msg, chat_history] + list(args)
             for response_chunk in chat_response(*backend_args):
                 yield {
-                    chat False
-        try:
-            backend_args = [user_msg, chat_history] + list(args)
-            for response_chunk in chat_response(*backend_args):
-                yield: response_chunk[0],
-                    dbg: response_chunk[1],
-                }
-        except GeneratorExit:
-            cancelled = True
-            print("Generation {
                     chat: response_chunk[0],
                     dbg: response_chunk[1],
                 }
         except GeneratorExit:
             cancelled = True
- cancelled by user.")
-            raise
-        except Exception as e:
-            print(f"An error occurred during generation: {            print("Generation cancelled by user.")
+            print("Generation cancelled by user.")
             raise
         except Exception as e:
             print(f"An error occurred during generation: {e}")
-e}")
             error_history = (chat_history or []) + [
                 {'role': 'user', 'content': user_msg},
-                           error_history = (chat_history or []) + [
-                {'role': 'user', 'content': {'role': 'assistant', 'content': f"**An error occurred:** {str user_msg},
-                {'role': 'assistant', 'content': f"(e)}"}
-            ]
-            yield {chat: error_history}
-        finally:
-            if not cancelled:
-                print("**An error occurred:** {str(e)}"}
+                {'role': 'assistant', 'content': f"**An error occurred:** {str(e)}"}
             ]
             yield {chat: error_history}
         finally:
             if not cancelled:
                 print("Resetting UI state.")
-                yield {
-                    txt: gr.update(interactive=True),
-                    submit_btn: gr.update(interactive=True),
-                   Resetting UI state.")
                 yield {
                     txt: gr.update(interactive=True),
                     submit_btn: gr.update(interactive=True),
                     cancel_btn: gr.update(visible=False),
                 }
 
-    def set_cancel_flag():
-        """Called by the cancel button, cancel_btn: gr.update(visible=False),
-                }
-
     def set_cancel_flag():
         """Called by the cancel button, sets the global event."""
         cancel_event.set()
         print("Cancellation signal sent.")
     
-    def reset_ui_after sets the global event."""
-        cancel_event.set()
-        print("Cancellation signal sent.")
-    
     def reset_ui_after_cancel():
-_cancel():
         """Reset UI components after cancellation."""
         cancel_event.clear()
         print("UI reset after cancellation.")
-        return {
-            txt: gr.update        """Reset UI components after cancellation."""
-        cancel_event.clear()
-        print("UI reset after cancellation.")
         return {
             txt: gr.update(interactive=True),
-            submit_btn: gr.update(interactive=True(interactive=True),
             submit_btn: gr.update(interactive=True),
             cancel_btn: gr.update(visible=False),
         }
 
-    # Event for submitting text via Enter key or Submit button
-    submit),
-            cancel_btn: gr.update(visible=False),
-        }
-
     # Event for submitting text via Enter key or Submit button
     submit_event = txt.submit(
-        fn=submit_and_event = txt.submit(
         fn=submit_and_manage_ui,
         inputs=chat_inputs,
         outputs=ui_components,
     )
-_manage_ui,
-        inputs=chat_inputs,
-        outputs=ui_components,
-    )
-    submit_    submit_btn.click(
-        fn=submit_and_manage_ui,
-        inputs=chat_inputs,
-        outputs=ui_components,
-   btn.click(
+    submit_btn.click(
         fn=submit_and_manage_ui,
         inputs=chat_inputs,
         outputs=ui_components,
     )
 
-    # Event )
-
     # Event for the "Cancel" button.
     cancel_btn.click(
         fn=set_cancel_flag,
         cancels=[submit_event]
     ).then(
-        fn=reset for the "Cancel" button.
-    cancel_btn.click(
-        fn=set_cancel_flag,
-        cancels=[submit_event]
-    ).then(
-        fn=reset_ui__ui_after_cancel,
-        outputs=ui_components
-    )
-
-    # Listeners forafter_cancel,
+        fn=reset_ui_after_cancel,
         outputs=ui_components
     )
 
     # Listeners for updating the duration estimate
-    duration updating the duration estimate
     duration_inputs = [model_dd, search_chk, mr, mc, max_tok, st]
-    for component in duration_inputs:
-        component_inputs = [model_dd, search_chk, mr, mc, max_tok, st]
     for component in duration_inputs:
         component.change(fn=update_duration_estimate, inputs=duration_inputs, outputs=duration_display)
 
-    # Toggle web search settings visibility
-    def toggle_search.change(fn=update_duration_estimate, inputs=duration_inputs, outputs=duration_display)
-
     # Toggle web search settings visibility
     def toggle_search_settings(enabled):
-       _settings(enabled):
         return gr.update(visible=enabled)
     
-    search_chk.change(
-        fn=lambda enabled: (update_default_prompt(enabled), gr.update return gr.update(visible=enabled)
-    
     search_chk.change(
         fn=lambda enabled: (update_default_prompt(enabled), gr.update(visible=enabled)),
         inputs=search_chk,
         outputs=[sys_prompt, search_settings]
     )
     
-    # Clear chat(visible=enabled)),
-        inputs=search_chk,
-        outputs=[sys_prompt, search_settings]
-    )
-    
-    # Clear chat action action
-    clr.click(fn=lambda: ([], "", ""), outputs=[chat, txt
+    # Clear chat action
     clr.click(fn=lambda: ([], "", ""), outputs=[chat, txt, dbg])
     
     demo.launch(share=True)