Reasoning on Unified Korean Benchmark Reasoning
77.2Ko-WinograndeQwen3-14B
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Qwen3-14BParameters=14B2026.01 | 77.2 | 75.4 | 6.4 | 64.5 | 48.8 | |
| Mi:dm 2.0 Base-instModel Size=Base, Instruction-tuned=true2026.01 | 75.1 | 73 | 8.6 | 52.9 | 44.8 | |
| Qwen3-4BParameters=4B2026.01 | 67.5 | 69.2 | 5.6 | 56.7 | 43.8 | |
| Exaone-3.5-7.8B-instParameters=7.8B, Instruction-tuned=true2026.01 | 64.6 | 60.3 | 8.6 | 49.7 | 39.5 | |
| Mi:dm 2.0 Mini-instModel Size=Mini, Instruction-tuned=true2026.01 | 61.7 | 64.5 | 7.7 | 39.9 | 37.4 | |
| Exaone-3.5-2.4B-instParameters=2.4B, Instruction-tuned=true2026.01 | 60.3 | 64.1 | 7.4 | 38.5 | 36.7 | |
| Llama-3.1-8B-instParameters=8B, Instruction-tuned=true2026.01 | 40.1 | 26 | 2.4 | 30.9 | 19.8 |