swtich agent and model leaderboard tabs and rename instances in the agent leaderboard

app.py

@@ -935,187 +935,166 @@ def build_app():
         """)
         
         with gr.Tabs():
-            with gr.Tab("🏆 Agent Leaderboard"):
-                # Visualization section
-
+            
+            
+            with gr.Tab("🤖 Model Leaderboard"):
                 with gr.Row():
                     gr.Markdown("""
                     **🎮 Welcome to LMGame Bench!** 
                     
-                    We welcome everyone to implement their own gaming agents by replacing our baseAgent in `customer_runner.py` and test them on our benchmark. Join the competition and see how your agent performs!
+                    We invite developers to implement their own gaming agents by replacing our `baseAgent` in `customer_runner.py` and evaluate them on our comprehensive benchmark. Visit our repository at https://github.com/lmgame-org/GamingAgent to get started and join the competition to see how your agent performs!
                     """, elem_classes="welcome-message")
-                    
+                
+                # Visualization section
                 with gr.Row():
                     gr.Markdown("### 📊 Data Visualization")
                 
                 # Detailed view visualization (single chart)
-                detailed_visualization = gr.Plot(
+                model_detailed_visualization = gr.Plot(
                     label="Performance Visualization",
                     visible=False,
                     elem_classes="visualization-container"
                 )
-                # with gr.Row():
-                #     # Calculate dynamic maximum based on total models
-                #     agent_max_models = get_total_model_count(rank_data)
-                #     top_n_slider = gr.Slider(
-                #         minimum=1,
-                #         maximum=agent_max_models,
-                #         step=1,
-                #         value=min(3, agent_max_models),
-                #         label=f"Number of Top Models to Display in All Views (max: {agent_max_models})",
-                #         elem_classes="top-n-slider"
-                #     )       
 
+                with gr.Row():
+                    # Calculate dynamic maximum based on total models
+                    model_max_models = get_total_model_count(model_rank_data)
+                    model_top_n_slider = gr.Slider(
+                        minimum=1,
+                        maximum=model_max_models,
+                        step=1,
+                        value=model_max_models,
+                        label=f"Number of Top Models to Display in All Views (max: {model_max_models})",
+                        elem_classes="top-n-slider"
+                    )
                 
+
                 
-                with gr.Column(visible=True) as overall_visualizations:
+                with gr.Column(visible=True) as model_overall_visualizations:
                     with gr.Tabs():
                         with gr.Tab("📈 Radar Chart"):
-                            
-                            radar_visualization = gr.Plot(
+                            model_radar_visualization = gr.Plot(
                                 label="Comparative Analysis (Radar Chart)",
                                 elem_classes="visualization-container"
                             )
                             gr.Markdown(
-                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*\n\n*🎮 GamingAgent - Our specialized gaming agents*",
+                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*",
                                     elem_classes="radar-tip"
                                 )
                         with gr.Tab("📊 Group Bar Chart"):
-                            group_bar_visualization = gr.Plot(
+                            model_group_bar_visualization = gr.Plot(
                                 label="Comparative Analysis (Group Bar Chart)",
                                 elem_classes="visualization-container"
                             )
                             gr.Markdown(
-                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*\n\n*🎮 GamingAgent - Our specialized gaming agents*",
+                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*",
                                     elem_classes="radar-tip"
                                 )
 
-                # Hidden placeholder for group bar visualization (to maintain code references)
-                # group_bar_visualization = gr.Plot(visible=False)
-
                 # Game selection section
                 with gr.Row():
                     gr.Markdown("### 🕹️ Game Selection")
                 with gr.Row():
-                    # with gr.Column(): # Commented out Super Mario BrosUI
-                    #     gr.Markdown("**🎮 Super Mario Bros**")
-                    #     mario_overall = gr.Checkbox(label="Super Mario BrosScore", value=True)
-                    #     mario_details = gr.Checkbox(label="Super Mario BrosDetails", value=False)
-                    with gr.Column(): # Added Super Mario BrosUI
+                    with gr.Column():
                         gr.Markdown("**🍄 Super Mario Bros**")
-                        mario_plan_overall = gr.Checkbox(label="Super Mario Bros Score", value=True)
-                        mario_plan_details = gr.Checkbox(label="Super Mario Bros Details", value=False)
-                    with gr.Column(): # Sokoban is now after mario_plan
+                        model_mario_plan_overall = gr.Checkbox(label="Super Mario Bros Score", value=True)
+                        model_mario_plan_details = gr.Checkbox(label="Super Mario Bros Details", value=False)
+                    with gr.Column():
                         gr.Markdown("**📦 Sokoban**")
-                        sokoban_overall = gr.Checkbox(label="Sokoban Score", value=True)
-                        sokoban_details = gr.Checkbox(label="Sokoban Details", value=False)
+                        model_sokoban_overall = gr.Checkbox(label="Sokoban Score", value=True)
+                        model_sokoban_details = gr.Checkbox(label="Sokoban Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🔢 2048**")
-                        _2048_overall = gr.Checkbox(label="2048 Score", value=True)
-                        _2048_details = gr.Checkbox(label="2048 Details", value=False)
+                        model_2048_overall = gr.Checkbox(label="2048 Score", value=True)
+                        model_2048_details = gr.Checkbox(label="2048 Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🍬 Candy Crush**")
-                        candy_overall = gr.Checkbox(label="Candy Crush Score", value=True)
-                        candy_details = gr.Checkbox(label="Candy Crush Details", value=False)
-                    # with gr.Column(): # Commented out Tetris(complete) UI
-                    #     gr.Markdown("**🎯 Tetris(complete)**")
-                    #     tetris_overall = gr.Checkbox(label="Tetris(complete) Score", value=True)
-                    #     tetris_details = gr.Checkbox(label="Tetris(complete) Details", value=False)
+                        model_candy_overall = gr.Checkbox(label="Candy Crush Score", value=True)
+                        model_candy_details = gr.Checkbox(label="Candy Crush Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🎯 Tetris**")
-                        tetris_plan_overall = gr.Checkbox(label="Tetris Score", value=True)
-                        tetris_plan_details = gr.Checkbox(label="Tetris Details", value=False)
+                        model_tetris_plan_overall = gr.Checkbox(label="Tetris Score", value=True)
+                        model_tetris_plan_details = gr.Checkbox(label="Tetris Details", value=False)
                     with gr.Column():
                         gr.Markdown("**⚖️ Ace Attorney**")
-                        ace_attorney_overall = gr.Checkbox(label="Ace Attorney Score", value=True)
-                        ace_attorney_details = gr.Checkbox(label="Ace Attorney Details", value=False)
+                        model_ace_attorney_overall = gr.Checkbox(label="Ace Attorney Score", value=True)
+                        model_ace_attorney_details = gr.Checkbox(label="Ace Attorney Details", value=False)
                 
                 # Controls
                 with gr.Row():
                     with gr.Column(scale=2):
                         gr.Markdown("**⏰ Time Tracker**")
-                        timeline = create_timeline_slider()
+                        model_timeline = create_timeline_slider()
                     with gr.Column(scale=1):
                         gr.Markdown("**🔄 Controls**")
-                        clear_btn = gr.Button("Reset Filters", variant="secondary")
+                        model_clear_btn = gr.Button("Reset Filters", variant="secondary")
                 
                 # Leaderboard table
                 with gr.Row():
                     gr.Markdown("### 📋 Detailed Results")
                 with gr.Row():
-                    gr.Markdown("*🎮 GamingAgent - Our specialized gaming agents*", elem_classes="radar-tip")
-                
-                # Welcome message for custom gaming agents
-                
-                # Add reference to Jupyter notebook
-                with gr.Row():
-                    gr.Markdown("*All data analysis can be replicated by checking [this Jupyter notebook](https://colab.research.google.com/drive/1CYFiJGm3EoBXXI8vICPVR82J9qrmmRvc#scrollTo=qft1Oald-21J)*")
+                    gr.Markdown("*💡 The slider above controls how many top models are shown in the radar chart, bar chart, and data table.*", elem_classes="radar-tip")
                 
-                # Get initial leaderboard dataframe (limited by default slider value for agent leaderboard)
-                initial_df = get_combined_leaderboard(rank_data, {
-                    # "Super Mario Bros": True, # Commented out
+                # Get initial leaderboard dataframe (limited by default slider value for model leaderboard)
+                model_initial_df = get_combined_leaderboard(model_rank_data, {
                     "Super Mario Bros": True,
                     "Sokoban": True,
                     "2048": True,
                     "Candy Crush": True,
-                    # "Tetris(complete)": True, # Commented out
                     "Tetris": True,
                     "Ace Attorney": True
-                }, limit_to_top_n=5)
+                }, limit_to_top_n=None)
                 
                 # Format the DataFrame for display
-                initial_display_df = prepare_dataframe_for_display(initial_df)
+                model_initial_display_df = prepare_dataframe_for_display(model_initial_df)
                 
-                # Custom column widths including row numbers
-                col_widths = ["40px"]  # Row number column width
-                col_widths.append("230px")  # Player column - reduced by 20px
-                col_widths.append("120px")  # Organization column
+                # Custom column widths including row numbers for model leaderboard
+                model_col_widths = ["40px"]  # Row number column width
+                model_col_widths.append("230px")  # Player column - reduced by 20px
+                model_col_widths.append("120px")  # Organization column
                 
                 # Check if there's an Avg Normalized Score column
-                if any('Avg Normalized' in col for col in initial_display_df.columns):
-                    col_widths.append("140px")  # Avg Normalized Score column - slightly wider
+                if any('Avg Normalized' in col for col in model_initial_display_df.columns):
+                    model_col_widths.append("140px")  # Avg Normalized Score column - slightly wider
                 
                 # Add game score columns
-                remaining_cols = len(initial_display_df.columns) - len(col_widths) + 1  # +1 because we subtracted row number column
+                remaining_cols = len(model_initial_display_df.columns) - len(model_col_widths) + 1  # +1 because we subtracted row number column
                 for _ in range(remaining_cols):
-                    col_widths.append("120px")
+                    model_col_widths.append("120px")
                 
                 # Create a standard DataFrame component with enhanced styling
                 with gr.Row():
-                    leaderboard_df = gr.DataFrame(
-                        value=initial_display_df,
+                    model_leaderboard_df = gr.DataFrame(
+                        value=model_initial_display_df,
                         interactive=True,
-                        elem_id="leaderboard-table",
+                        elem_id="model-leaderboard-table",
                         elem_classes="table-container",
                         wrap=True,
                         show_row_numbers=True,
                         show_fullscreen_button=True,
                         line_breaks=True,
-                        max_height=1000,  # Set a larger fixed height
+                        max_height=1000,
                         show_search="search",
-                        column_widths=col_widths
+                        column_widths=model_col_widths
                     )
                 
                 # Add the score note below the table
                 with gr.Row():
-                    score_note = add_score_note()
+                    model_score_note = add_score_note()
                 
-                # List of all checkboxes, including Super Mario Bros
-                checkbox_list = [
-                    # mario_overall, mario_details, # Commented out
-                    mario_plan_overall, mario_plan_details,
-                    sokoban_overall, sokoban_details,
-                    _2048_overall, _2048_details,
-                    candy_overall, candy_details,
-                    # tetris_overall, tetris_details, # Commented out
-                    tetris_plan_overall, tetris_plan_details,
-                    ace_attorney_overall, ace_attorney_details
+                # List of all checkboxes for model leaderboard
+                model_checkbox_list = [
+                    model_mario_plan_overall, model_mario_plan_details,
+                    model_sokoban_overall, model_sokoban_details,
+                    model_2048_overall, model_2048_details,
+                    model_candy_overall, model_candy_details,
+                    model_tetris_plan_overall, model_tetris_plan_details,
+                    model_ace_attorney_overall, model_ace_attorney_details
                 ]
                 
                 # Update visualizations when checkboxes change
-                def update_visualizations(*checkbox_states):
+                def update_model_visualizations(*checkbox_states):
                     # Check if any details checkbox is selected
-                    # Adjusted indices due to addition of Super Mario 
                     is_details_view = any([
                         checkbox_states[1], # Mario Plan details
                         checkbox_states[3], # Sokoban details
@@ -1127,192 +1106,253 @@ def build_app():
                     
                     # Update visibility of visualization blocks
                     return {
-                        detailed_visualization: gr.update(visible=is_details_view),
-                        overall_visualizations: gr.update(visible=not is_details_view)
+                        model_detailed_visualization: gr.update(visible=is_details_view),
+                        model_overall_visualizations: gr.update(visible=not is_details_view)
                     }
                 
                 # Add change event to all checkboxes
-                for checkbox in checkbox_list:
+                for checkbox in model_checkbox_list:
                     checkbox.change(
-                        update_visualizations,
-                        inputs=checkbox_list,
-                        outputs=[detailed_visualization, overall_visualizations]
+                        update_model_visualizations,
+                        inputs=model_checkbox_list,
+                        outputs=[model_detailed_visualization, model_overall_visualizations]
                     )
                 
                 # Update leaderboard and visualizations when checkboxes change
-                for checkbox in checkbox_list:
+                for checkbox in model_checkbox_list:
                     checkbox.change(
-                        lambda *args: update_leaderboard(*args, top_n=5, data_source=rank_data),
-                        inputs=checkbox_list,
+                        lambda *args: update_leaderboard(*args, data_source=model_rank_data),
+                        inputs=model_checkbox_list + [model_top_n_slider],
                         outputs=[
-                            leaderboard_df,
-                            detailed_visualization,
-                            radar_visualization,
-                            group_bar_visualization
-                        ] + checkbox_list
+                            model_leaderboard_df,
+                            model_detailed_visualization,
+                            model_radar_visualization,
+                            model_group_bar_visualization
+                        ] + model_checkbox_list
                     )
                 
+                # Update when model top_n_slider changes
+                model_top_n_slider.change(
+                    lambda *args: update_leaderboard(*args, data_source=model_rank_data),
+                    inputs=model_checkbox_list + [model_top_n_slider],
+                    outputs=[
+                        model_leaderboard_df,
+                        model_detailed_visualization,
+                        model_radar_visualization,
+                        model_group_bar_visualization
+                    ] + model_checkbox_list
+                )
+                
                 # Update when clear button is clicked
-                clear_btn.click(
-                    lambda: clear_filters(top_n=5, data_source=rank_data),
-                    inputs=[],
+                model_clear_btn.click(
+                    lambda *args: clear_filters(*args, data_source=model_rank_data),
+                    inputs=[model_top_n_slider],
                     outputs=[
-                        leaderboard_df,
-                        detailed_visualization,
-                        radar_visualization,
-                        group_bar_visualization
-                    ] + checkbox_list
+                        model_leaderboard_df,
+                        model_detailed_visualization,
+                        model_radar_visualization,
+                        model_group_bar_visualization
+                    ] + model_checkbox_list
                 )
                 
-                # Initialize the agent leaderboard (with top 5 limit)
+                # Initialize the model leaderboard (with all models shown by default)
                 demo.load(
-                    lambda: clear_filters(top_n=5, data_source=rank_data),
+                    lambda: clear_filters(top_n=get_total_model_count(model_rank_data), data_source=model_rank_data),
                     inputs=[],
                     outputs=[
-                        leaderboard_df,
-                        detailed_visualization,
-                        radar_visualization,
-                        group_bar_visualization
-                    ] + checkbox_list
+                        model_leaderboard_df,
+                        model_detailed_visualization,
+                        model_radar_visualization,
+                        model_group_bar_visualization
+                    ] + model_checkbox_list
                 )
             
-            with gr.Tab("🤖 Model Leaderboard"):
+            with gr.Tab("🏆 Agent Leaderboard"):
                 # Visualization section
+
+                with gr.Row():
+                    # Calculate dynamic maximum based on total models
+                    agent_max_models = get_total_model_count(rank_data)
+                    agent_top_n_slider = gr.Slider(
+                        minimum=1,
+                        maximum=agent_max_models,
+                        step=1,
+                        value=5,
+                        label=f"Number of Top Models to Display in All Views (max: {agent_max_models})",
+                        elem_classes="top-n-slider"
+                    )
+                    
                 with gr.Row():
                     gr.Markdown("### 📊 Data Visualization")
                 
                 # Detailed view visualization (single chart)
-                model_detailed_visualization = gr.Plot(
+                detailed_visualization = gr.Plot(
                     label="Performance Visualization",
                     visible=False,
                     elem_classes="visualization-container"
                 )
+                # with gr.Row():
+                #     # Calculate dynamic maximum based on total models
+                #     agent_max_models = get_total_model_count(rank_data)
+                #     top_n_slider = gr.Slider(
+                #         minimum=1,
+                #         maximum=agent_max_models,
+                #         step=1,
+                #         value=min(3, agent_max_models),
+                #         label=f"Number of Top Models to Display in All Views (max: {agent_max_models})",
+                #         elem_classes="top-n-slider"
+                #     )       
 
-                with gr.Row():
-                    # Calculate dynamic maximum based on total models
-                    model_max_models = get_total_model_count(model_rank_data)
-                    model_top_n_slider = gr.Slider(
-                        minimum=1,
-                        maximum=model_max_models,
-                        step=1,
-                        value=model_max_models,
-                        label=f"Number of Top Models to Display in All Views (max: {model_max_models})",
-                        elem_classes="top-n-slider"
-                    )
                 
-
                 
-                with gr.Column(visible=True) as model_overall_visualizations:
+                with gr.Column(visible=True) as overall_visualizations:
                     with gr.Tabs():
                         with gr.Tab("📈 Radar Chart"):
-                            model_radar_visualization = gr.Plot(
+                            
+                            radar_visualization = gr.Plot(
                                 label="Comparative Analysis (Radar Chart)",
                                 elem_classes="visualization-container"
                             )
                             gr.Markdown(
-                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*",
+                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*\n\n*🎮 Model Name (GamingAgent) - Our specialized gaming agents*",
                                     elem_classes="radar-tip"
                                 )
                         with gr.Tab("📊 Group Bar Chart"):
-                            model_group_bar_visualization = gr.Plot(
+                            group_bar_visualization = gr.Plot(
                                 label="Comparative Analysis (Group Bar Chart)",
                                 elem_classes="visualization-container"
                             )
                             gr.Markdown(
-                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*",
+                                    "*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*\n\n*🎮 Model Name (GamingAgent) - Our specialized gaming agents*",
                                     elem_classes="radar-tip"
                                 )
 
+                # Hidden placeholder for group bar visualization (to maintain code references)
+                # group_bar_visualization = gr.Plot(visible=False)
+
                 # Game selection section
                 with gr.Row():
                     gr.Markdown("### 🕹️ Game Selection")
                 with gr.Row():
-                    with gr.Column():
+                    # with gr.Column(): # Commented out Super Mario BrosUI
+                    #     gr.Markdown("**🎮 Super Mario Bros**")
+                    #     mario_overall = gr.Checkbox(label="Super Mario BrosScore", value=True)
+                    #     mario_details = gr.Checkbox(label="Super Mario BrosDetails", value=False)
+                    with gr.Column(): # Added Super Mario BrosUI
                         gr.Markdown("**🍄 Super Mario Bros**")
-                        model_mario_plan_overall = gr.Checkbox(label="Super Mario Bros Score", value=True)
-                        model_mario_plan_details = gr.Checkbox(label="Super Mario Bros Details", value=False)
-                    with gr.Column():
+                        mario_plan_overall = gr.Checkbox(label="Super Mario Bros Score", value=True)
+                        mario_plan_details = gr.Checkbox(label="Super Mario Bros Details", value=False)
+                    with gr.Column(): # Sokoban is now after mario_plan
                         gr.Markdown("**📦 Sokoban**")
-                        model_sokoban_overall = gr.Checkbox(label="Sokoban Score", value=True)
-                        model_sokoban_details = gr.Checkbox(label="Sokoban Details", value=False)
+                        sokoban_overall = gr.Checkbox(label="Sokoban Score", value=True)
+                        sokoban_details = gr.Checkbox(label="Sokoban Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🔢 2048**")
-                        model_2048_overall = gr.Checkbox(label="2048 Score", value=True)
-                        model_2048_details = gr.Checkbox(label="2048 Details", value=False)
+                        _2048_overall = gr.Checkbox(label="2048 Score", value=True)
+                        _2048_details = gr.Checkbox(label="2048 Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🍬 Candy Crush**")
-                        model_candy_overall = gr.Checkbox(label="Candy Crush Score", value=True)
-                        model_candy_details = gr.Checkbox(label="Candy Crush Details", value=False)
+                        candy_overall = gr.Checkbox(label="Candy Crush Score", value=True)
+                        candy_details = gr.Checkbox(label="Candy Crush Details", value=False)
+                    # with gr.Column(): # Commented out Tetris(complete) UI
+                    #     gr.Markdown("**🎯 Tetris(complete)**")
+                    #     tetris_overall = gr.Checkbox(label="Tetris(complete) Score", value=True)
+                    #     tetris_details = gr.Checkbox(label="Tetris(complete) Details", value=False)
                     with gr.Column():
                         gr.Markdown("**🎯 Tetris**")
-                        model_tetris_plan_overall = gr.Checkbox(label="Tetris Score", value=True)
-                        model_tetris_plan_details = gr.Checkbox(label="Tetris Details", value=False)
+                        tetris_plan_overall = gr.Checkbox(label="Tetris Score", value=True)
+                        tetris_plan_details = gr.Checkbox(label="Tetris Details", value=False)
                     with gr.Column():
                         gr.Markdown("**⚖️ Ace Attorney**")
-                        model_ace_attorney_overall = gr.Checkbox(label="Ace Attorney Score", value=True)
-                        model_ace_attorney_details = gr.Checkbox(label="Ace Attorney Details", value=False)
+                        ace_attorney_overall = gr.Checkbox(label="Ace Attorney Score", value=True)
+                        ace_attorney_details = gr.Checkbox(label="Ace Attorney Details", value=False)
                 
                 # Controls
                 with gr.Row():
                     with gr.Column(scale=2):
                         gr.Markdown("**⏰ Time Tracker**")
-                        model_timeline = create_timeline_slider()
+                        timeline = create_timeline_slider()
                     with gr.Column(scale=1):
                         gr.Markdown("**🔄 Controls**")
-                        model_clear_btn = gr.Button("Reset Filters", variant="secondary")
+                        clear_btn = gr.Button("Reset Filters", variant="secondary")
                 
                 # Leaderboard table
                 with gr.Row():
                     gr.Markdown("### 📋 Detailed Results")
                 with gr.Row():
-                    gr.Markdown("*💡 The slider above controls how many top models are shown in the radar chart, bar chart, and data table.*", elem_classes="radar-tip")
+                    gr.Markdown("*🎮 Model Name (GamingAgent) - Our specialized gaming agents*", elem_classes="radar-tip")
                 
-                # Get initial leaderboard dataframe (limited by default slider value for model leaderboard)
-                model_initial_df = get_combined_leaderboard(model_rank_data, {
+                # Welcome message for custom gaming agents
+                
+                # Add reference to Jupyter notebook
+                with gr.Row():
+                    gr.Markdown("*All data analysis can be replicated by checking [this Jupyter notebook](https://colab.research.google.com/drive/1CYFiJGm3EoBXXI8vICPVR82J9qrmmRvc#scrollTo=qft1Oald-21J)*")
+                
+                # Get initial leaderboard dataframe (limited by default slider value for agent leaderboard)
+                initial_df = get_combined_leaderboard(rank_data, {
+                    # "Super Mario Bros": True, # Commented out
                     "Super Mario Bros": True,
                     "Sokoban": True,
                     "2048": True,
                     "Candy Crush": True,
+                    # "Tetris(complete)": True, # Commented out
                     "Tetris": True,
                     "Ace Attorney": True
-                }, limit_to_top_n=None)
+                }, limit_to_top_n=5)
                 
                 # Format the DataFrame for display
-                model_initial_display_df = prepare_dataframe_for_display(model_initial_df)
+                initial_display_df = prepare_dataframe_for_display(initial_df)
+                
+                # Custom column widths including row numbers
+                col_widths = ["40px"]  # Row number column width
+                col_widths.append("230px")  # Player column - reduced by 20px
+                col_widths.append("120px")  # Organization column
+                
+                # Check if there's an Avg Normalized Score column
+                if any('Avg Normalized' in col for col in initial_display_df.columns):
+                    col_widths.append("140px")  # Avg Normalized Score column - slightly wider
+                
+                # Add game score columns
+                remaining_cols = len(initial_display_df.columns) - len(col_widths) + 1  # +1 because we subtracted row number column
+                for _ in range(remaining_cols):
+                    col_widths.append("120px")
                 
                 # Create a standard DataFrame component with enhanced styling
                 with gr.Row():
-                    model_leaderboard_df = gr.DataFrame(
-                        value=model_initial_display_df,
+                    leaderboard_df = gr.DataFrame(
+                        value=initial_display_df,
                         interactive=True,
-                        elem_id="model-leaderboard-table",
+                        elem_id="leaderboard-table",
                         elem_classes="table-container",
                         wrap=True,
                         show_row_numbers=True,
                         show_fullscreen_button=True,
                         line_breaks=True,
-                        max_height=1000,
+                        max_height=1000,  # Set a larger fixed height
                         show_search="search",
                         column_widths=col_widths
                     )
                 
                 # Add the score note below the table
                 with gr.Row():
-                    model_score_note = add_score_note()
+                    score_note = add_score_note()
                 
-                # List of all checkboxes for model leaderboard
-                model_checkbox_list = [
-                    model_mario_plan_overall, model_mario_plan_details,
-                    model_sokoban_overall, model_sokoban_details,
-                    model_2048_overall, model_2048_details,
-                    model_candy_overall, model_candy_details,
-                    model_tetris_plan_overall, model_tetris_plan_details,
-                    model_ace_attorney_overall, model_ace_attorney_details
+                # List of all checkboxes, including Super Mario Bros
+                checkbox_list = [
+                    # mario_overall, mario_details, # Commented out
+                    mario_plan_overall, mario_plan_details,
+                    sokoban_overall, sokoban_details,
+                    _2048_overall, _2048_details,
+                    candy_overall, candy_details,
+                    # tetris_overall, tetris_details, # Commented out
+                    tetris_plan_overall, tetris_plan_details,
+                    ace_attorney_overall, ace_attorney_details
                 ]
                 
                 # Update visualizations when checkboxes change
-                def update_model_visualizations(*checkbox_states):
+                def update_visualizations(*checkbox_states):
                     # Check if any details checkbox is selected
+                    # Adjusted indices due to addition of Super Mario 
                     is_details_view = any([
                         checkbox_states[1], # Mario Plan details
                         checkbox_states[3], # Sokoban details
@@ -1324,65 +1364,65 @@ def build_app():
                     
                     # Update visibility of visualization blocks
                     return {
-                        model_detailed_visualization: gr.update(visible=is_details_view),
-                        model_overall_visualizations: gr.update(visible=not is_details_view)
+                        detailed_visualization: gr.update(visible=is_details_view),
+                        overall_visualizations: gr.update(visible=not is_details_view)
                     }
                 
                 # Add change event to all checkboxes
-                for checkbox in model_checkbox_list:
+                for checkbox in checkbox_list:
                     checkbox.change(
-                        update_model_visualizations,
-                        inputs=model_checkbox_list,
-                        outputs=[model_detailed_visualization, model_overall_visualizations]
+                        update_visualizations,
+                        inputs=checkbox_list,
+                        outputs=[detailed_visualization, overall_visualizations]
                     )
                 
                 # Update leaderboard and visualizations when checkboxes change
-                for checkbox in model_checkbox_list:
+                for checkbox in checkbox_list:
                     checkbox.change(
-                        lambda *args: update_leaderboard(*args, data_source=model_rank_data),
-                        inputs=model_checkbox_list + [model_top_n_slider],
+                        lambda *args: update_leaderboard(*args, data_source=rank_data),
+                        inputs=checkbox_list + [agent_top_n_slider],
                         outputs=[
-                            model_leaderboard_df,
-                            model_detailed_visualization,
-                            model_radar_visualization,
-                            model_group_bar_visualization
-                        ] + model_checkbox_list
+                            leaderboard_df,
+                            detailed_visualization,
+                            radar_visualization,
+                            group_bar_visualization
+                        ] + checkbox_list
                     )
                 
-                # Update when model top_n_slider changes
-                model_top_n_slider.change(
-                    lambda *args: update_leaderboard(*args, data_source=model_rank_data),
-                    inputs=model_checkbox_list + [model_top_n_slider],
+                # Update when agent top_n_slider changes
+                agent_top_n_slider.change(
+                    lambda *args: update_leaderboard(*args, data_source=rank_data),
+                    inputs=checkbox_list + [agent_top_n_slider],
                     outputs=[
-                        model_leaderboard_df,
-                        model_detailed_visualization,
-                        model_radar_visualization,
-                        model_group_bar_visualization
-                    ] + model_checkbox_list
+                        leaderboard_df,
+                        detailed_visualization,
+                        radar_visualization,
+                        group_bar_visualization
+                    ] + checkbox_list
                 )
                 
                 # Update when clear button is clicked
-                model_clear_btn.click(
-                    lambda *args: clear_filters(*args, data_source=model_rank_data),
-                    inputs=[model_top_n_slider],
+                clear_btn.click(
+                    lambda *args: clear_filters(*args, data_source=rank_data),
+                    inputs=[agent_top_n_slider],
                     outputs=[
-                        model_leaderboard_df,
-                        model_detailed_visualization,
-                        model_radar_visualization,
-                        model_group_bar_visualization
-                    ] + model_checkbox_list
+                        leaderboard_df,
+                        detailed_visualization,
+                        radar_visualization,
+                        group_bar_visualization
+                    ] + checkbox_list
                 )
                 
-                # Initialize the model leaderboard (with all models shown by default)
+                # Initialize the agent leaderboard (with top 5 limit)
                 demo.load(
-                    lambda: clear_filters(top_n=get_total_model_count(model_rank_data), data_source=model_rank_data),
+                    lambda: clear_filters(top_n=5, data_source=rank_data),
                     inputs=[],
                     outputs=[
-                        model_leaderboard_df,
-                        model_detailed_visualization,
-                        model_radar_visualization,
-                        model_group_bar_visualization
-                    ] + model_checkbox_list
+                        leaderboard_df,
+                        detailed_visualization,
+                        radar_visualization,
+                        group_bar_visualization
+                    ] + checkbox_list
                 )
             
             with gr.Tab("🎥 Gallery"):

assets/model_color.json

@@ -27,31 +27,31 @@
     "llama-4-maverick-17b-128e-instruct-fp8": "#8E24AA",
     "qwen3-235B-A22B-fp8": "#6A1B9A",
     "random (x30)": "#9E9E9E",
-    "🎮 GamingAgent (claude-3-7-sonnet-20250219)": "#4A90E2",
-    "🎮 GamingAgent (claude-3-5-haiku-20241022)": "#7FB5E6",
-    "🎮 GamingAgent (claude-3-5-sonnet-20241022)": "#1A4C7C",
-    "🎮 GamingAgent (claude-opus-4-20250514)": "#3A80D2",
-    "🎮 GamingAgent (claude-sonnet-4-20250514)": "#5A9FE2",
-    "🎮 GamingAgent (gemini-2.0-flash)": "#FF4081",
-    "🎮 GamingAgent (gemini-2.0-flash-thinking-exp-1219)": "#C2185B",
-    "🎮 GamingAgent (gemini-2.5-pro-exp-03-25)": "#FF80AB",
-    "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)": "#F06292",
-    "🎮 GamingAgent (gemini-2.5-flash-preview-05-20)": "#F8BBD9",
-    "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)": "#AD1457",
-    "🎮 GamingAgent (gemini-2.5-pro-preview-06-05)": "#EC407A",
-    "🎮 GamingAgent (gpt-4o-2024-11-20)": "#00BFA5",
-    "🎮 GamingAgent (gpt-4.5-preview-2025-02-27)": "#00796B",
-    "🎮 GamingAgent (gpt-4.1-2025-04-14)": "#00897B",
-    "🎮 GamingAgent (o1-2024-12-17)": "#4DB6AC",
-    "🎮 GamingAgent (o1-mini-2024-09-12)": "#26A69A",
-    "🎮 GamingAgent (o3-mini-2025-01-31(medium))": "#80CBC4",
-    "🎮 GamingAgent (o3-2025-04-16)": "#26C6DA",
-    "🎮 GamingAgent (o4-mini-2025-04-16)": "#00ACC1",
-    "🎮 GamingAgent (grok-3-beta)": "#FF7043",
-    "🎮 GamingAgent (grok-3-mini-beta)": "#FF8A65",
-    "🎮 GamingAgent (deepseek-v3)": "#FFC107",
-    "🎮 GamingAgent (deepseek-r1-0120)": "#FFA000",
-    "🎮 GamingAgent (deepseek-r1-0528)": "#FFB300",
-    "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)": "#8E24AA",
-    "🎮 GamingAgent (qwen3-235B-A22B-fp8)": "#6A1B9A"
+    "🎮 claude-3-7-sonnet-20250219 (GamingAgent)": "#4A90E2",
+    "🎮 claude-3-5-haiku-20241022 (GamingAgent)": "#7FB5E6",
+    "🎮 claude-3-5-sonnet-20241022 (GamingAgent)": "#1A4C7C",
+    "🎮 claude-opus-4-20250514 (GamingAgent)": "#3A80D2",
+    "🎮 claude-sonnet-4-20250514 (GamingAgent)": "#5A9FE2",
+    "🎮 gemini-2.0-flash (GamingAgent)": "#FF4081",
+    "🎮 gemini-2.0-flash-thinking-exp-1219 (GamingAgent)": "#C2185B",
+    "🎮 gemini-2.5-pro-exp-03-25 (GamingAgent)": "#FF80AB",
+    "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)": "#F06292",
+    "🎮 gemini-2.5-flash-preview-05-20 (GamingAgent)": "#F8BBD9",
+    "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)": "#AD1457",
+    "🎮 gemini-2.5-pro-preview-06-05 (GamingAgent)": "#EC407A",
+    "🎮 gpt-4o-2024-11-20 (GamingAgent)": "#00BFA5",
+    "🎮 gpt-4.5-preview-2025-02-27 (GamingAgent)": "#00796B",
+    "🎮 gpt-4.1-2025-04-14 (GamingAgent)": "#00897B",
+    "🎮 o1-2024-12-17 (GamingAgent)": "#4DB6AC",
+    "🎮 o1-mini-2024-09-12 (GamingAgent)": "#26A69A",
+    "🎮 o3-mini-2025-01-31(medium) (GamingAgent)": "#80CBC4",
+    "🎮 o3-2025-04-16 (GamingAgent)": "#26C6DA",
+    "🎮 o4-mini-2025-04-16 (GamingAgent)": "#00ACC1",
+    "🎮 grok-3-beta (GamingAgent)": "#FF7043",
+    "🎮 grok-3-mini-beta (GamingAgent)": "#FF8A65",
+    "🎮 deepseek-v3 (GamingAgent)": "#FFC107",
+    "🎮 deepseek-r1-0120 (GamingAgent)": "#FFA000",
+    "🎮 deepseek-r1-0528 (GamingAgent)": "#FFB300",
+    "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)": "#8E24AA",
+    "🎮 qwen3-235B-A22B-fp8 (GamingAgent)": "#6A1B9A"
 }

rank_data_03_25_2025.json

@@ -3,61 +3,61 @@
         "runs": 3,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 1267.7,
                 "detail_data": "709,1532,1562",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 1418.7,
                 "detail_data": "2015,709,1532",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 1385.0,
                 "detail_data": "1672,1266,1247",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 1498.3,
                 "detail_data": "1561,1271,1663",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 1468.7,
                 "detail_data": "898,2008,1500",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 2126.3,
                 "detail_data": "1531,722,4126",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 2047.3,
                 "detail_data": "2017,2590,1535",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 855,
                 "detail_data": "855",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 3445,
                 "detail_data": "3445",
                 "progress": "1-1"
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 1448.0,
                 "detail_data": "1525,1263,1556",
                 "progress": "1-1"
@@ -74,79 +74,79 @@
         "runs": 3,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 1914.67,
                 "details": "1352,2860,1532",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 2624,
                 "details": "2560,3224,2088",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0120)",
+                "model": "🎮 deepseek-r1-0120 (GamingAgent)",
                 "score": 1873.33,
                 "details": "700,1240,3680",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 1697.33,
                 "details": "1304,1316,2472",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 3586.67,
                 "details": "5300,2400,3060",
                 "highest_tail": 512
             },
             {
-                "model": "🎮 GamingAgent (grok-3-mini-beta)",
+                "model": "🎮 grok-3-mini-beta (GamingAgent)",
                 "score": 4036,
                 "details": "6412,2492,3204",
                 "highest_tail": 512
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 1586.67,
                 "details": "1404,1272,2084",
                 "highest_tail": 128
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 1656,
                 "details": "1156,2664,1148",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 1656,
                 "details": "1604,1284,2080",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 7580,
                 "details": "7580",
                 "highest_tail": 512
             },
             {
-                "model": "🎮 GamingAgent (o1-mini-2024-09-12)",
+                "model": "🎮 o1-mini-2024-09-12 (GamingAgent)",
                 "score": 2757.33,
                 "details": "3132,2004,3136",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 7120,
                 "details": "7120",
                 "highest_tail": 512
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 4432.0,
                 "details": "4928,5456,2912",
                 "highest_tail": 512
@@ -158,25 +158,25 @@
                 "highest_tail": 128
             },
             {
-                "model": "🎮 GamingAgent (claude-opus-4-20250514)",
+                "model": "🎮 claude-opus-4-20250514 (GamingAgent)",
                 "score": 3036.0,
                 "details": "3036.0",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (claude-sonnet-4-20250514)",
+                "model": "🎮 claude-sonnet-4-20250514 (GamingAgent)",
                 "score": 3136,
                 "details": "2148,2360,4900",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0528)",
+                "model": "🎮 deepseek-r1-0528 (GamingAgent)",
                 "score": 3330.0,
                 "details": "3260,3400",
                 "highest_tail": 256
             },
             {
-                "model": "🎮 GamingAgent (qwen3-235B-A22B-fp8)",
+                "model": "🎮 qwen3-235B-A22B-fp8 (GamingAgent)",
                 "score": 2144.0,
                 "details": "1436,2556,2440",
                 "highest_tail": 256
@@ -187,67 +187,67 @@
         "runs": 3,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 14.7,
                 "details": "16,14,14"
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 16.3,
                 "details": "19,15,15"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0120)",
+                "model": "🎮 deepseek-r1-0120 (GamingAgent)",
                 "score": 14.3,
                 "details": "15,14,14"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 16.3,
                 "details": "20,14,15"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 23.3,
                 "details": "23,23,24"
             },
             {
-                "model": "🎮 GamingAgent (grok-3-mini-beta)",
+                "model": "🎮 grok-3-mini-beta (GamingAgent)",
                 "score": 21.3,
                 "details": "20,15,29"
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 10.3,
                 "details": "9,10,12"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 13.7,
                 "details": "13,14,14"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 14,
                 "details": "18,11,13"
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 35,
                 "details": "35"
             },
             {
-                "model": "🎮 GamingAgent (o1-mini-2024-09-12)",
+                "model": "🎮 o1-mini-2024-09-12 (GamingAgent)",
                 "score": 11.7,
                 "details": "11,11,13"
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 42,
                 "details": "42"
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 25.3,
                 "details": "22,35,19"
             },
@@ -257,22 +257,22 @@
                 "details": ""
             },
             {
-                "model": "🎮 GamingAgent (claude-opus-4-20250514)",
+                "model": "🎮 claude-opus-4-20250514 (GamingAgent)",
                 "score": 20,
                 "details": "17,18,25"
             },
             {
-                "model": "🎮 GamingAgent (claude-sonnet-4-20250514)",
+                "model": "🎮 claude-sonnet-4-20250514 (GamingAgent)",
                 "score": 19.33,
                 "details": "20,17,21"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0528)",
+                "model": "🎮 deepseek-r1-0528 (GamingAgent)",
                 "score": 33.67,
                 "details": "26,34,41"
             },
             {
-                "model": "🎮 GamingAgent (qwen3-235B-A22B-fp8)",
+                "model": "🎮 qwen3-235B-A22B-fp8 (GamingAgent)",
                 "score": 11.67,
                 "details": "13,14,8"
             }
@@ -282,67 +282,67 @@
         "runs": 3,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 106,
                 "details": "92,165,61"
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 484,
                 "details": "535,428,489"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0120)",
+                "model": "🎮 deepseek-r1-0120 (GamingAgent)",
                 "score": 447.3,
                 "details": "409,436,497"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 334.7,
                 "details": "259,372,373"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 416.3,
                 "details": "411,414,424"
             },
             {
-                "model": "🎮 GamingAgent (grok-3-mini-beta)",
+                "model": "🎮 grok-3-mini-beta (GamingAgent)",
                 "score": 254,
                 "details": "299,332,131"
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 128.7,
                 "details": "67,139,180"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 182,
                 "details": "163,215,168"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 147.3,
                 "details": "131,104,207"
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 159,
                 "details": "159"
             },
             {
-                "model": "🎮 GamingAgent (o1-mini-2024-09-12)",
+                "model": "🎮 o1-mini-2024-09-12 (GamingAgent)",
                 "score": 48,
                 "details": "21,86,37"
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 647,
                 "details": "647"
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 487.3,
                 "details": "259,591,612"
             },
@@ -352,22 +352,22 @@
                 "details": ""
             },
             {
-                "model": "🎮 GamingAgent (claude-opus-4-20250514)",
+                "model": "🎮 claude-opus-4-20250514 (GamingAgent)",
                 "score": 464,
                 "details": "593,406,393"
             },
             {
-                "model": "🎮 GamingAgent (claude-sonnet-4-20250514)",
+                "model": "🎮 claude-sonnet-4-20250514 (GamingAgent)",
                 "score": 478.33,
                 "details": "545,468,422"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0528)",
+                "model": "🎮 deepseek-r1-0528 (GamingAgent)",
                 "score": 491.67,
                 "details": "464,463,548"
             },
             {
-                "model": "🎮 GamingAgent (qwen3-235B-A22B-fp8)",
+                "model": "🎮 qwen3-235B-A22B-fp8 (GamingAgent)",
                 "score": 363.33,
                 "details": "365,372,353"
             }
@@ -377,79 +377,79 @@
         "runs": 3,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 0,
                 "detail_box_on_target": "0,0,0",
                 "cracked_levels": "0,0,0"
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 2.33,
                 "detail_box_on_target": "2,4,1",
                 "cracked_levels": "1,2,0"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0120)",
+                "model": "🎮 deepseek-r1-0120 (GamingAgent)",
                 "score": 1.33,
                 "detail_box_on_target": "2,0,2",
                 "cracked_levels": "1,0,1"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 1.67,
                 "detail_box_on_target": "3,0,2",
                 "cracked_levels": "2,0,1"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 4.33,
                 "detail_box_on_target": "4,4,5",
                 "cracked_levels": "2,2,3"
             },
             {
-                "model": "🎮 GamingAgent (grok-3-mini-beta)",
+                "model": "🎮 grok-3-mini-beta (GamingAgent)",
                 "score": 5.67,
                 "detail_box_on_target": "5,6,6",
                 "cracked_levels": "3,3,3"
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 0,
                 "detail_box_on_target": "0,0,0",
                 "cracked_levels": "0,0,0"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 0,
                 "detail_box_on_target": "0,0,0",
                 "cracked_levels": "0,0,0"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 0,
                 "detail_box_on_target": "0,0,0",
                 "cracked_levels": "0,0,0"
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 2.33,
                 "detail_box_on_target": "2,2,3",
                 "cracked_levels": "1,1,2"
             },
             {
-                "model": "🎮 GamingAgent (o1-mini-2024-09-12)",
+                "model": "🎮 o1-mini-2024-09-12 (GamingAgent)",
                 "score": 1.33,
                 "detail_box_on_target": "1,2,1",
                 "cracked_levels": "0,1,0"
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 8,
                 "detail_box_on_target": "10,6",
                 "cracked_levels": "5,3"
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 5.33,
                 "detail_box_on_target": "4,6,6",
                 "cracked_levels": "2,2,3"
@@ -461,22 +461,22 @@
                 "cracked_levels": "0,0,0"
             },
             {
-                "model": "🎮 GamingAgent (claude-opus-4-20250514)",
+                "model": "🎮 claude-opus-4-20250514 (GamingAgent)",
                 "score": 4,
                 "details": "4,4,4"
             },
             {
-                "model": "🎮 GamingAgent (claude-sonnet-4-20250514)",
+                "model": "🎮 claude-sonnet-4-20250514 (GamingAgent)",
                 "score": 3,
                 "details": "2,2,5"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0528)",
+                "model": "🎮 deepseek-r1-0528 (GamingAgent)",
                 "score": 4.67,
                 "details": "4,4,6"
             },
             {
-                "model": "🎮 GamingAgent (qwen3-235B-A22B-fp8)",
+                "model": "🎮 qwen3-235B-A22B-fp8 (GamingAgent)",
                 "score": 2.33,
                 "details": "1,2,4"
             }
@@ -486,79 +486,79 @@
         "runs": 1,
         "results": [
             {
-                "model": "🎮 GamingAgent (claude-3-5-sonnet-20241022)",
+                "model": "🎮 claude-3-5-sonnet-20241022 (GamingAgent)",
                 "score": 2,
                 "progress": "1:2/5",
                 "evaluator result": "1/3"
             },
             {
-                "model": "🎮 GamingAgent (claude-3-7-sonnet-20250219)",
+                "model": "🎮 claude-3-7-sonnet-20250219 (GamingAgent)",
                 "score": 7,
                 "progress": "2:2/9",
                 "evaluator result": "5/11"
             },
             {
-                "model": "🎮 GamingAgent (deepseek-r1-0120)",
+                "model": "🎮 deepseek-r1-0120 (GamingAgent)",
                 "score": 0,
                 "progress": "0",
                 "evaluator result": "1/5"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-04-17)",
+                "model": "🎮 gemini-2.5-flash-preview-04-17 (GamingAgent)",
                 "score": 4,
                 "progress": "1:4/5",
                 "evaluator result": "1/7"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-pro-preview-05-06)",
+                "model": "🎮 gemini-2.5-pro-preview-05-06 (GamingAgent)",
                 "score": 7,
                 "progress": "2:2/9",
                 "evaluator result": "2/3"
             },
             {
-                "model": "🎮 GamingAgent (grok-3-mini-beta)",
+                "model": "🎮 grok-3-mini-beta (GamingAgent)",
                 "score": 0,
                 "progress": "0",
                 "evaluator result": "0"
             },
             {
-                "model": "🎮 GamingAgent (llama-4-maverick-17b-128e-instruct-fp8)",
+                "model": "🎮 llama-4-maverick-17b-128e-instruct-fp8 (GamingAgent)",
                 "score": 0,
                 "progress": "0",
                 "evaluator result": "0"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4.1-2025-04-14)",
+                "model": "🎮 gpt-4.1-2025-04-14 (GamingAgent)",
                 "score": 2,
                 "progress": "1:2/5",
                 "evaluator result": "2/3"
             },
             {
-                "model": "🎮 GamingAgent (gpt-4o-2024-11-20)",
+                "model": "🎮 gpt-4o-2024-11-20 (GamingAgent)",
                 "score": 0,
                 "progress": "0",
                 "evaluator result": "0"
             },
             {
-                "model": "🎮 GamingAgent (o1-2024-12-17)",
+                "model": "🎮 o1-2024-12-17 (GamingAgent)",
                 "score": 16,
                 "progress": "3: 2/8",
                 "evaluator result": "6/11"
             },
             {
-                "model": "🎮 GamingAgent (o1-mini-2024-09-12)",
+                "model": "🎮 o1-mini-2024-09-12 (GamingAgent)",
                 "score": 0,
                 "progress": "0",
                 "evaluator result": "1/5"
             },
             {
-                "model": "🎮 GamingAgent (o3-2025-04-16)",
+                "model": "🎮 o3-2025-04-16 (GamingAgent)",
                 "score": 16,
                 "progress": "3: 2/8",
                 "evaluator result": "1/2"
             },
             {
-                "model": "🎮 GamingAgent (o4-mini-2025-04-16)",
+                "model": "🎮 o4-mini-2025-04-16 (GamingAgent)",
                 "score": 4,
                 "progress": "1:4/5",
                 "evaluator result": "2/5"
@@ -570,17 +570,17 @@
                 "evaluator result": "0"
             },
             {
-                "model": "🎮 GamingAgent (claude-opus-4-20250514)",
+                "model": "🎮 claude-opus-4-20250514 (GamingAgent)",
                 "score": 6,
                 "details": "6"
             },
             {
-                "model": "🎮 GamingAgent (claude-sonnet-4-20250514)",
+                "model": "🎮 claude-sonnet-4-20250514 (GamingAgent)",
                 "score": 3.67,
                 "details": "3,4,4"
             },
             {
-                "model": "🎮 GamingAgent (gemini-2.5-flash-preview-05-20)",
+                "model": "🎮 gemini-2.5-flash-preview-05-20 (GamingAgent)",
                 "score": 4.33,
                 "details": "3,4,6"
             }