Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agent Performance Evaluation on MCP-Bench

46.8Task Fulfillment

ReAct

30.78434.94239.143.258May 9, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.05
46.836.641.531.340.519.841.736.430.136.1
2026.05
38.538.742.547.836.920.638.645.228.837.5
2026.05
34.547.555.358.433.327.54156.930.442.8
2026.05
33.742.645.240.429.626.838.242.828.236.4
2026.05
32.85159.66133.630.941.960.332.344.8
2026.05
31.626.633.648.420.520.229.14120.330.1
2026.05
31.447.253.165.927.129.939.359.528.542.4