List your company

Tool-use Planning on ToolBench G3-Instruction

0.9368Win RateGPT4 TOPGUN
0.5345280.6389640.74340.847836Feb 15, 2024

Evaluation results

MethodLinks
GPT4 TOPGUNReference=GPT4 ReACT2024.020.9368
GPT4 TOPGUNReference=T.LLAMA REACT2024.020.9323
GPT4 TOPGUNReference=ChatGPT REACT2024.020.9005
GPT4 TOPGUNReference=T.LLAMA DFSDT2024.020.8947
GPT4 TOPGUNReference=T.LLAMA DFSDT+Ret2024.020.8782
GPT4 TOPGUNReference=ChatGPT DFSDT2024.020.8526
GPT4 DFSDTReference=ChatGPT REACT2024.020.84
GPT4 TOPGUNReference=GPT4 DFSDT2024.020.7925
GPT4 ReACTReference=ChatGPT REACT2024.020.78
T.LLAMA DFSDT+RetReference=ChatGPT REACT2024.020.73
T.LLAMA DFSDTReference=ChatGPT REACT2024.020.69
ChatGPT DFSDTReference=ChatGPT REACT2024.020.69
T.LLAMA REACTReference=ChatGPT REACT2024.020.55