Mathematical Reasoning on AIME 24 (mean@32 accuracy)
74.7Mean@32 AccuracyQwen3-30B-A3B-Instruct-2507
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-30B-A3B-Instruct-2507Role=Teacher Model2026.06 | 74.7 | |
| IW-OPDStudent=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 57.5 | |
| IW-OPDStudent=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 56.2 | |
| OPDStudent=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 55.3 | |
| OPDStudent=Qwen3-4B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 51.2 | |
| IW-OPDStudent=Qwen3-1.7B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 35.5 | |
| OPDStudent=Qwen3-1.7B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 34.6 | |
| IW-OPDStudent=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 33 | |
| OPDStudent=Qwen3-1.7B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 30.5 | |
| BaseStudent=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 23.1 | |
| BaseStudent=Qwen3-1.7B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 13.4 | |
| IW-OPDStudent=Qwen3-0.6B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 11.5 | |
| OPDStudent=Qwen3-0.6B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 11 | |
| IW-OPDStudent=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 7.8 | |
| OPDStudent=Qwen3-0.6B, Training Step=10, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 6.2 | |
| BaseStudent=Qwen3-0.6B, Teacher Model=Qwen3-30B-A3B-Instruct-25072026.06 | 1.5 |