Mathematical Reasoning on Math_500 CodaSet OOD (test)
90.33Accuracy (%)DeepSeek-V3.2-Exp
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-V3.2-Exp2026.05 | 90.33 | 3.5 | |
| DeepSeek-V3.12026.05 | 89.86 | 5.6 | |
| Qwen3-235B2026.05 | 85.98 | 5.3 | |
| DecoRBase Model=DeepSeek-V3.12026.05 | 85.51 | 3.1 | |
| gpt-oss-120b2026.05 | 83.18 | 1.9 | |
| EmbedLLM2026.05 | 82.71 | 5.1 | |
| MODEL-SAT2026.05 | 81.98 | 2.9 | |
| Kimi-K2-Instruct2026.05 | 81.31 | 15.2 | |
| RouterDC2026.05 | 80.64 | 3.9 | |
| DecoRBase Model=gpt-oss-120b2026.05 | 80.37 | 1.4 | |
| Random Router2026.05 | 80.18 | 4 | |
| gemma-3-27b-it2026.05 | 78.04 | 1.3 | |
| LLM Router2026.05 | 77.57 | 1.2 | |
| KNN Router2026.05 | 77.1 | 1.1 | |
| Mistral-3.2-24B2026.05 | 76.64 | 1.4 | |
| gemma-3-12b-it2026.05 | 75.7 | 1 |