Multi-agent cooperation sustainability simulation on GOVSIM default experiment
53.3Survival RateGPT-4o
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GPT-4oWeight Access=Closed-Weights2024.04 | 53.3 | 9.3 | 66 | 55 | 94.4 | 10.8 | |
| Claude-3 OpusWeight Access=Closed-Weights2024.04 | 46.7 | 6.9 | 58.5 | 48.8 | 91.4 | 21 | |
| GPT-4-turboWeight Access=Closed-Weights2024.04 | 40 | 6.6 | 62.4 | 52 | 93.6 | 15.7 | |
| Qwen-110BWeight Access=Open-Weights2024.04 | 20 | 4.5 | 36.3 | 30.3 | 89.6 | 47 | |
| GPT-4Weight Access=Closed-Weights2024.04 | 6.7 | 3.9 | 31.5 | 26.2 | 91.4 | 27.1 | |
| Llama-3-8BWeight Access=Open-Weights2024.04 | 0 | 1 | 20 | 16.7 | 57.3 | 20 | |
| Llama-3-70BWeight Access=Open-Weights2024.04 | 0 | 1 | 20 | 16.7 | 90.7 | 38.7 | |
| Mistral-7BWeight Access=Open-Weights2024.04 | 0 | 1 | 20 | 16.7 | 82.6 | 37.3 | |
| Mixtral-8x7BWeight Access=Open-Weights2024.04 | 0 | 1.1 | 20.1 | 16.7 | 75 | 33.3 | |
| Qwen-72BWeight Access=Open-Weights2024.04 | 0 | 1.8 | 24 | 20 | 83.9 | 32.4 | |
| Claude-3 HaikuWeight Access=Closed-Weights2024.04 | 0 | 1 | 20 | 16.7 | 91 | 35.7 | |
| Claude-3 SonnetWeight Access=Closed-Weights2024.04 | 0 | 1.3 | 20.5 | 17.1 | 84.4 | 32 | |
| GPT-3.5Weight Access=Closed-Weights2024.04 | 0 | 1.1 | 20.3 | 16.9 | 91.2 | 35.3 |