Sustainability Test on GOVSIM Pasture
80Survival RateClaude-3 Opus
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Claude-3 OpusWeight access=Closed-Weights2024.04 | 80 | 10.2 | 99.24 | 82.7 | 98.23 | 9.86 | |
| GPT-4oWeight access=Closed-Weights2024.04 | 20 | 6.6 | 57.92 | 48.27 | 94.7 | 24.61 | |
| Llama-2-7BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 46.48 | 17.4 | |
| Llama-2-13BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 49.6 | 14.29 | |
| Llama-2-70BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 77.84 | 48 | |
| Llama-3-8BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 61.44 | 24.29 | |
| Llama-3-70BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 92.4 | 40.52 | |
| Mistral-7BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 88.64 | 42.61 | |
| Mixtral-8x7BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 80.16 | 34.33 | |
| Qwen-72BWeight access=Open-Weights2024.04 | 0 | 1 | 20 | 16.67 | 86 | 40.28 | |
| Qwen-110BWeight access=Open-Weights2024.04 | 0 | 3.2 | 27.76 | 23.13 | 86.52 | 56.55 | |
| Claude-3 HaikuWeight access=Closed-Weights2024.04 | 0 | 1 | 20 | 16.67 | 87.52 | 35.71 | |
| Claude-3 SonnetWeight access=Closed-Weights2024.04 | 0 | 1 | 20 | 16.67 | 87.6 | 34.29 | |
| GPT-3.5Weight access=Closed-Weights2024.04 | 0 | 1 | 20 | 16.67 | 90.88 | 35.71 | |
| GPT-4Weight access=Closed-Weights2024.04 | 0 | 1.8 | 21.92 | 18.27 | 93.18 | 37.84 | |
| GPT-4-turboWeight access=Closed-Weights2024.04 | 0 | 2 | 23.12 | 19.27 | 91.63 | 35.11 |