Multi-task Reasoning on BigBench Hard
31.1ScoreGPT-4o
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-4oDate=2024-11-202025.08 | 31.1 | 1 | |
| Sonnet 3.5Date=2024-10-222025.08 | 30.4 | 2 | |
| Haiku 3.52025.08 | 30.1 | 3 | |
| Gemini 2.0 Flash2025.08 | 28.7 | 4 | |
| Haiku 32025.08 | 8.2 | 5 |
| Method | Links | ||
|---|---|---|---|
| GPT-4oDate=2024-11-202025.08 | 31.1 | 1 | |
| Sonnet 3.5Date=2024-10-222025.08 | 30.4 | 2 | |
| Haiku 3.52025.08 | 30.1 | 3 | |
| Gemini 2.0 Flash2025.08 | 28.7 | 4 | |
| Haiku 32025.08 | 8.2 | 5 |