Agricultural Agent Planning and Execution on AgriAgent Full set
100CoverageAgriAgent
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AgriAgentModel=Qwen3-32B2026.01 | 100 | 78.6 | 73.1 | 78.3 | 77.6 | 79.6 | 76.7 | 70.9 | |
| AgriAgentModel=Qwen3-235B2026.01 | 100 | 100 | 100 | — | 87.2 | — | 81.9 | 79.8 | |
| AgriAgentModel=Qwen3-8B2026.01 | 98.7 | 49.7 | 50.5 | 78.6 | 68.9 | 76.8 | 66.1 | 57.6 | |
| AgriAgentModel=AgriGPT-8B2026.01 | 98.1 | 57.2 | 58.6 | 70 | 72.9 | 79 | 62.2 | 63.9 | |
| AgriAgentModel=Qwen2.5-14B2026.01 | 96.9 | 74.1 | 80.9 | 50 | 77.5 | 78.6 | 67.5 | 67.9 | |
| AgriAgent (w/o feedback)Model=Qwen3-8B2026.01 | 96.7 | 66.2 | 66.4 | 78.6 | 66.9 | 77.1 | 62.8 | 58 | |
| AgriAgent (w/o feedback)Model=Qwen3-32B2026.01 | 96.6 | 74.1 | 69 | 73.6 | 75.2 | 78.8 | 72.7 | 68.4 | |
| AgriAgentModel=DeepSeek-v3.12026.01 | 95 | 94.7 | 95 | — | 80.9 | — | 73.8 | 71.5 | |
| AgriAgentModel=Qwen2.5-7B2026.01 | 94.4 | 71.9 | 75.7 | 55.5 | 71.9 | 70.2 | 61.9 | 60.5 | |
| AgriAgent (w/o feedback)Model=AgriGPT-8B2026.01 | 91.9 | 62.8 | 63.8 | 72.2 | 68.8 | 77.6 | 63.9 | 63.1 | |
| AgriAgentModel=LLaMA3-8B2026.01 | 90.9 | 67.3 | 76.6 | 35.8 | 55.5 | 74.8 | 47.3 | 47.9 | |
| AgriAgent (w/o feedback)Model=Qwen2.5-14B2026.01 | 69.1 | 51 | 53.1 | 48.9 | 53.5 | 78 | 48.5 | 46.1 | |
| reactModel=Qwen2.5-14B2026.01 | 62.4 | 9.2 | 4.3 | 46 | 43.1 | 74.6 | 20.3 | 31.7 | |
| plan-and-executeModel=Qwen2.5-7B2026.01 | 59.3 | 1.3 | 10 | 42 | 39.6 | 69.3 | 18 | 28.9 | |
| plan-and-executeModel=Qwen2.5-14B2026.01 | 56.7 | 3.1 | 8.4 | 32 | 36.6 | 76.9 | 19.4 | 30.3 | |
| AgriAgent (w/o feedback)Model=Qwen2.5-7B2026.01 | 55.9 | 37.3 | 39.2 | 54.3 | 37.5 | 70.5 | 33.9 | 31.2 | |
| reactModel=AgriGPT-8B2026.01 | 53.8 | 0.8 | 2.6 | 62 | 26.7 | 71.7 | 14.2 | 22.6 | |
| AgriAgent (w/o feedback)Model=LLaMA3-8B2026.01 | 49.5 | 31.9 | 39.3 | 31.5 | 21.8 | 74.6 | 21.8 | 19.8 | |
| plan-and-executeModel=Qwen3-32B2026.01 | 49.4 | 8.5 | 11.2 | 61 | 35.4 | 78.1 | 24.4 | 30.3 | |
| plan-and-executeModel=AgriGPT-8B2026.01 | 48 | 6.3 | 8.8 | 65 | 29 | 77.5 | 20.9 | 27 | |
| reactModel=Qwen3-8B2026.01 | 43.7 | 0.3 | 0.3 | 65 | 21.5 | 73.8 | 12.3 | 19.6 | |
| plan-and-executeModel=Qwen3-8B2026.01 | 43.3 | 4.8 | 6.7 | 61 | 26.4 | 76.1 | 18.1 | 23.5 | |
| reactModel=Qwen3-32B2026.01 | 17.8 | 0.7 | 1.3 | 60 | 7.9 | 70 | 4.6 | 7.9 | |
| reactModel=Qwen2.5-7B2026.01 | 12.1 | 0 | 0 | 35 | 7 | 66.7 | 2.2 | 4.7 | |
| reactModel=LLaMA3-8B2026.01 | 6.3 | 0.3 | 0 | 1 | 3.4 | 42.4 | 1.7 | 3 | |
| plan-and-executeModel=LLaMA3-8B2026.01 | 0.4 | 0 | 0 | 2.9 | 0.2 | 41.6 | 0.1 | 0.2 |