Comprehensive Reasoning on Mathematical and General Reasoning Suite Combined
46.7Overall AccuracyWIST
Evaluation Results
| Method | Links | |
|---|---|---|
| WISTBackbone=Qwen3-8B-Base2026.03 | 46.7 | |
| SPICEBackbone=Qwen3-8B-Base2026.03 | 46 | |
| R-ZeroBackbone=Qwen3-8B-Base2026.03 | 45.5 | |
| WISTBackbone=Qwen3-4B-Base2026.03 | 43.1 | |
| Base ModelBackbone=Qwen3-8B-Base2026.03 | 42.1 | |
| SPICEBackbone=Qwen3-4B-Base2026.03 | 41.8 | |
| R-ZeroBackbone=Qwen3-4B-Base2026.03 | 40.6 | |
| Base ModelBackbone=Qwen3-4B-Base2026.03 | 33.3 | |
| WISTBackbone=OctoThinker-8B-Hybrid-Base2026.03 | 32.6 | |
| SPICEBackbone=OctoThinker-8B-Hybrid-Base2026.03 | 31.4 | |
| R-ZeroBackbone=OctoThinker-8B-Hybrid-Base2026.03 | 30.6 | |
| SPICEBackbone=OctoThinker-3B-Hybrid-Base2026.03 | 25.2 | |
| WISTBackbone=OctoThinker-3B-Hybrid-Base2026.03 | 24.5 | |
| Base ModelBackbone=OctoThinker-8B-Hybrid-Base2026.03 | 22.9 | |
| R-ZeroBackbone=OctoThinker-3B-Hybrid-Base2026.03 | 22.5 | |
| Base ModelBackbone=OctoThinker-3B-Hybrid-Base2026.03 | 13.2 |