BIG-Bench Hard — Results
| Rank | Model | Score |
|---|---|---|
| 1 | ling-3-0-flash-base | 89.17 |
| 2 | deepseek-v4-1-flash | 86.1 |
| 3 | ling-3-0-tiny-base | 80.64 |
| 4 | claude-3-5-sonnet | 0.931 |
| 5 | gemini-1-5-pro | 0.892 |
| 6 | claude-3-opus | 0.868 |
| 7 | gemini-1-5-flash | 0.855 |
| 8 | phi-3-5-moe-instruct | 0.791 |
| 9 | phi-4-mini | 0.704 |
| 10 | phi-3-5-mini-instruct | 0.69 |
BIG-Bench Hard — Evaluations
1 phaseActive
No evaluations yet for this benchmark.