Science Reasoning on ARC-C Korean (test)
90.5AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o2026.06 | 90.5 | |
| Qwen3 235B A22B2026.06 | 90.3 | |
| LuckyStar 111B2026.06 | 89.2 | |
| Claude 3.7 Sonnet2026.06 | 89 | |
| Command A2026.06 | 88.8 | |
| LuckyStar 111BQuantization=4-bit2026.06 | 88.6 |