ToolBench

ToolBench is a benchmark that evaluates the ability of large language models to utilize external tools through a stable and large-scale framework, incorporating a virtual API server and a systematic evaluation approach.

Papers using ToolBench (3)

UTFix: Change Aware Unit Test Repairing using LLM2025 · 3 citesAnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls2024 · 1 citesStableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models2024