Sustainability Test on Fishery Sustainability (test)
100Survival RateGPT-4-turbo
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GPT-4-turboWeight Accessibility=Closed-Weights2024.04 | 100 | 12 | 108.8 | 90.67 | 98.05 | 0.51 | |
| GPT-4oWeight Accessibility=Closed-Weights2024.04 | 100 | 12 | 71.36 | 59.47 | 98.03 | 0.35 | |
| Claude-3 OpusWeight Accessibility=Closed-Weights2024.04 | 60 | 9.6 | 56.28 | 46.9 | 94.57 | 18.79 | |
| Qwen-110BWeight Accessibility=Open-Weights2024.04 | 40 | 6.6 | 49.04 | 40.87 | 88.65 | 28.51 | |
| GPT-4Weight Accessibility=Closed-Weights2024.04 | 20 | 5.2 | 32.52 | 27.1 | 92.02 | 22.43 | |
| Llama-2-7BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 74.32 | 45.08 | |
| Llama-2-13BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 88.72 | 35.48 | |
| Llama-2-70BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 100 | 59.72 | |
| Llama-3-8BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 67.6 | 21.43 | |
| Llama-3-70BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 88.16 | 39.4 | |
| Mistral-7BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 85.76 | 40.13 | |
| Mixtral-8x7BWeight Accessibility=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 85.52 | 40.87 | |
| Qwen-72BWeight Accessibility=Open-Weights2024.04 | 0 | 3.4 | 32 | 26.67 | 84.9 | 25.45 | |
| Claude-3 HaikuWeight Accessibility=Closed-Weights2024.04 | 0 | 1 | 20 | 16.67 | 97.44 | 35.71 | |
| Claude-3 SonnetWeight Accessibility=Closed-Weights2024.04 | 0 | 2 | 21.56 | 17.97 | 93.64 | 33.17 | |
| GPT-3.5Weight Accessibility=Closed-Weights2024.04 | 0 | 1.4 | 20.8 | 17.33 | 91.69 | 32.16 |