Step Prediction on Pen-Strategist (test)
82.87AccuracyStep Model
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Step Model2026.05 | 82.87 | 80 | |
| Gemini 2.5 Flash2026.05 | 64.52 | 68 | |
| Claude 4.5 sonnet2026.05 | 63.44 | 66 | |
| GPT 4.12026.05 | 62.78 | 65 | |
| Gemini 2.0 Flash2026.05 | 61.29 | 67 | |
| GPT-5-mini2026.05 | 59.66 | 65 | |
| Claude 3 Haiku2026.05 | 55.43 | 53 | |
| GPT-4o-mini2026.05 | 54.83 | 54 | |
| LLaMA-3.1-8B2026.05 | 51.09 | 54 | |
| GPT-3.5-turbo2026.05 | 41.76 | 45 |