List your company

Multi-task Reasoning on BigBench Hard

31.1ScoreGPT-4o
7.28413.46719.6525.833Aug 8, 2025

Evaluation results

MethodLinks
GPT-4oDate=2024-11-202025.0831.11
Sonnet 3.5Date=2024-10-222025.0830.42
Haiku 3.52025.0830.13
Gemini 2.0 Flash2025.0828.74
Haiku 32025.088.25