List your company
ResearchFormerly WizWand

Tool-use Planning on ToolBench Average over all sets

86.54Win RateGPT4 TOPGUN
45.418456.094266.7777.4458Feb 15, 2024
Updated 8mo ago

Evaluation results

MethodLinks
GPT4 TOPGUNReference=ChatGPT REACT2024.0286.54
GPT4 TOPGUNReference=T.LLAMA REACT2024.0284.61
GPT4 TOPGUNReference=GPT4 ReACT2024.0280.27
GPT4 TOPGUNReference=T.LLAMA DFSDT+Ret2024.0279.44
GPT4 TOPGUNReference=ChatGPT DFSDT2024.0278.71
GPT4 TOPGUNReference=GPT4 DFSDT2024.0278.59
GPT4 TOPGUNReference=T.LLAMA DFSDT2024.0278.44
GPT4 DFSDTReference=ChatGPT REACT2024.0270.4
ChatGPT DFSDTReference=ChatGPT REACT2024.0264.3
GPT4 ReACTReference=ChatGPT REACT2024.0264
T.LLAMA DFSDT+RetReference=ChatGPT REACT2024.0263.1
T.LLAMA DFSDTReference=ChatGPT REACT2024.0260
T.LLAMA REACTReference=ChatGPT REACT2024.0247