Mathematical Reasoning on MATH 500 (Korean)
95.6Pass@1 AccuracyLuckyStar 111B 4-bit
Evaluation results
| Method | Links | |
|---|---|---|
| LuckyStar 111B 4-bitQuantization=4-bit, Context Window=32k2026.06 | 95.6 | |
| LuckyStar 111BContext Window=32k2026.06 | 93.6 | |
| Claude 3.7 SonnetContext Window=32k2026.06 | 92.8 | |
| Qwen3 235B A22BContext Window=32k2026.06 | 92.4 | |
| GPT-4o (11/20)Context Window=32k2026.06 | 76.4 | |
| Command AContext Window=32k2026.06 | 76.2 |