Behavioral Prediction on Woodward-like Behavioral Prediction Non-animate
94AccuracyClaude Opus 4.6
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Opus 4.62025.03 | 94 | |
| Gemini 3.1 Pro2025.03 | 90 | |
| Qwen 3.5 Plus2025.03 | 88 | |
| Humans2025.03 | 81 | |
| GPT-4o2025.03 | 60 | |
| GPT-5.22025.03 | 58 | |
| Claude 3.5 Sonnet2025.03 | 50 | |
| Qwen VL Max2025.03 | 49 | |
| Gemini 3.1 Flash2025.03 | 47 |