Problem Solving and Unsolvability Detection on HamPath
100Solvable AccuracyGemini-3
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini-3Model Scale=32025.12 | 100 | 90 | 95 | |
| Deepseek-V3.2-RModel Scale=V3.2-R2025.12 | 84 | 96 | 90 | |
| Qwen3-4B + UnsolvableRLModel Scale=4B, Training Protocol=UnsolvableRL2025.12 | 55 | 96.5 | 75.8 | |
| GPT-5.1-LowModel Scale=5.1-Low2025.12 | 50 | 86 | 68 | |
| Qwen3-4B InstructModel Scale=4B, Training Protocol=Instruct2025.12 | 37 | 56.5 | 46.8 | |
| Qwen3-1.7B + UnsolvableRLModel Scale=1.7B, Training Protocol=UnsolvableRL2025.12 | 27 | 85 | 56 | |
| Qwen3-1.7B InstructModel Scale=1.7B, Training Protocol=Instruct2025.12 | 14 | 33 | 23.5 |