Overall Evaluation on Principle-based evaluation dataset
8.41AverageClaude 4.5-Haiku
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude 4.5-HaikuAccess Type=API-based2025.12 | 8.41 | |
| Claude 4.5-SonnetAccess Type=API-based2025.12 | 8.41 | |
| GPT-5.1Access Type=API-based2025.12 | 8.36 | |
| Claude 3.5-HaikuAccess Type=API-based2025.12 | 8.33 | |
| GPT-4.1Access Type=API-based2025.12 | 8.3 | |
| Qwen3-8BAccess Type=Open-source2025.12 | 8.23 | |
| Qwen3-4BAccess Type=Open-source2025.12 | 8.21 | |
| GPT-5Access Type=API-based2025.12 | 8.21 | |
| Qwen3-1.7BAccess Type=Open-source2025.12 | 8.03 | |
| Qwen2.5-7BAccess Type=Open-source2025.12 | 7.97 | |
| Llama-3.1-8BAccess Type=Open-source2025.12 | 7.86 | |
| Llama-3.2-3BAccess Type=Open-source2025.12 | 7.81 |