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 21 Jul 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 30 Jul 2026.")
                 
             with gr.Tabs(elem_id="inner-tabs", elem_classes="tabs-small") as tabs:
                 with gr.TabItem("Leaderboard"):

report_generation.jsonl

@@ -96,3 +96,4 @@
 {"Model": "Google/Gemini-3.6-Flash", "Category": "Closed-source Instruct", "Overall": 55.0, "Physics": 38.0, "Chemistry": 68.7, "Finance": 40.3, "Consulting": 73.1, "Extraction": 51.1, "Reasoning": 56.0, "Style": 66.8, "Response Characters": 7698, "Input Tokens": 479, "Output Tokens": 2405, "Cost": 3.0}
 {"Model": "Google/Gemini-3.5-Flash-Lite (Thinking)", "Category": "Closed-source Reasoning", "Overall": 51.6, "Physics": 37.3, "Chemistry": 64.9, "Finance": 34.4, "Consulting": 69.8, "Extraction": 45.7, "Reasoning": 53.1, "Style": 62.1, "Response Characters": 5726, "Input Tokens": 479, "Output Tokens": 10959, "Cost": 4.41}
 {"Model": "Google/Gemini-3.5-Flash-Lite", "Category": "Closed-source Instruct", "Overall": 45.4, "Physics": 34.2, "Chemistry": 49.6, "Finance": 26.8, "Consulting": 71.2, "Extraction": 38.0, "Reasoning": 47.0, "Style": 62.6, "Response Characters": 6377, "Input Tokens": 479, "Output Tokens": 1918, "Cost": 0.79}
+{"Model": "Anthropic/claude-opus-5 (Thinking)", "Category": "Closed-source Reasoning", "Overall": 66.2, "Physics": 48.2, "Chemistry": 80.5, "Finance": 61.5, "Consulting": 74.6, "Extraction": 60.6, "Reasoning": 69.9, "Style": 72.0, "Response Characters": 10470, "Input Tokens": 716, "Output Tokens": 15599, "Cost": 62.97}