Mathematical Reasoning on HMMT Nov
81.67AccuracyQwen3-Next-80B-A3B
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-Next-80B-A3BParameters=80B2026.02 | 81.67 | |
| Nanbeige4.1-3BParameters=3B2026.02 | 77.92 | |
| Qwen3-30B-A3B-2507Parameters=30B2026.02 | 71.25 | |
| Qwen3-4B-2507Parameters=4B2026.02 | 68.33 | |
| Nanbeige4-3B-2511Parameters=3B2026.02 | 66.67 | |
| Qwen3-32BParameters=32B2026.02 | 57.08 | |
| SPSBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.04 | 43.33 | |
| GSPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.04 | 40 | |
| DeepSeek-R1-Distill-Qwen-1.5BType=Base Model2026.04 | 36.67 | |
| GRPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.04 | 36.67 | |
| DAPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.04 | 26.67 |