Upload 2 files

app.py

@@ -111,7 +111,7 @@ with gr.Blocks(theme=theme) as app:
         with gr.TabItem("Report Generation"):
             with gr.Row():
                 with gr.Column(scale=7):
-                    gr.Markdown("Report Generation Leaderboard: LLMs generate reports with just the prompt, which are then evaluated by gpt-oss-120b (mixed) judge with the lite dataset (160 samples) \nEvaluation and cost estimation last performed on 5 Mar 2026.")
+                    gr.Markdown("Report Generation Leaderboard: LLMs generate reports with just the prompt, which are then evaluated by gpt-oss-120b (mixed) judge with the lite dataset (160 samples) \nEvaluation and cost estimation last performed on 21 Mar 2026.")
                 
             with gr.Tabs(elem_id="inner-tabs", elem_classes="tabs-small") as tabs:
                 with gr.TabItem("Leaderboard"):

report_generation.jsonl

@@ -65,3 +65,4 @@
 {"Model": "OpenAI/GPT-5.4 (xhigh)", "Category": "Closed-source Reasoning", "Overall": 56.8, "Physics": 41.7, "Chemistry": 70.7, "Finance": 47.4, "Consulting": 67.6, "Extraction": 47.1, "Reasoning": 58.0, "Style": 71.7, "Response Characters": 6246, "Input Tokens": 2516, "Output Tokens": 25806, "Cost": 62.94}
 {"Model": "OpenAI/GPT-5.3-Codex (xhigh)", "Category": "Closed-source Reasoning", "Overall": 57.2, "Physics": 39.5, "Chemistry": 67.4, "Finance": 51.9, "Consulting": 69.9, "Extraction": 51.4, "Reasoning": 58.2, "Style": 75.5, "Response Characters": 3596, "Input Tokens": 467, "Output Tokens": 15414, "Cost": 34.66}
 {"Model": "OpenAI/GPT-5.3-Chat", "Category": "Closed-source Instruct", "Overall": 40.7, "Physics": 36.1, "Chemistry": 53.5, "Finance": 18.6, "Consulting": 54.6, "Extraction": 29.6, "Reasoning": 41.4, "Style": 59.2, "Response Characters": 3108, "Input Tokens": 467, "Output Tokens": 1328, "Cost": 3.11}
+{"Model": "xAI/grok-4.20 Beta (Thinking)", "Category": "Closed-source Reasoning", "Overall": 43.8, "Physics": 34.4, "Chemistry": 64.8, "Finance": 19.6, "Consulting": 56.2, "Extraction": 32.9, "Reasoning": 43.0, "Style": 65.6, "Response Characters": 4003, "Input Tokens": 566, "Output Tokens": 9507, "Cost": 9.31}