Algorithmic Program Execution on Pattern1 Algorithmic Task n=100 (test)
100AccuracyGemini-3.1-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3.1-ProProtocol=LLM-PV, Tools=with tools2025.10 | 100 | |
| GPT-5-ThinkingProtocol=LLM-PV, Tools=with tools2025.10 | 100 | |
| Gemini-3.1-ProProtocol=ICL, Tools=with tools2025.10 | 83 | |
| Llama-3.2-1BProtocol=FT2025.10 | 61.5 | |
| XGBoost2025.10 | 50.6 | |
| Random Forest2025.10 | 50.5 |