Tool Execution on ToolBench Tool Gen
52.2SoPRToolOmni
Evaluation Results
| Method | Links | |
|---|---|---|
| ToolOmniEvaluation Setting=End-to-End, Interaction Budget=62026.04 | 52.2 | |
| GPT-3.5Evaluation Setting=End-to-End, Interaction Budget=62026.04 | 50.6 | |
| ToolLlama-v2Evaluation Setting=End-to-End, Interaction Budget=62026.04 | 39.2 | |
| ToolGen (7B)Evaluation Setting=End-to-End, Interaction Budget=62026.04 | 33.2 |