AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls

TL;DR

AnyTool is a hierarchical, self-reflective language model agent that efficiently utilizes a large set of APIs to improve query resolution, demonstrating significant performance gains over existing methods.

Contribution

We present AnyTool, a novel hierarchical agent with self-reflection capabilities that leverages GPT-4's function calling to utilize extensive APIs without external training.

Findings

Outperforms strong baselines like ToolLLM by +35.4% on ToolBench

Introduces a revised evaluation protocol for more realistic assessment

Demonstrates effective use of 16,000+ APIs for query solving

Abstract

We introduce AnyTool, a large language model agent designed to revolutionize the utilization of a vast array of tools in addressing user queries. We utilize over 16,000 APIs from Rapid API, operating under the assumption that a subset of these APIs could potentially resolve the queries. AnyTool primarily incorporates three elements: an API retriever with a hierarchical structure, a solver aimed at resolving user queries using a selected set of API candidates, and a self-reflection mechanism, which re-activates AnyTool if the initial solution proves impracticable. AnyTool is powered by the function calling feature of GPT-4, eliminating the need for training external modules. We also revisit the evaluation protocol introduced by previous works and identify a limitation in this protocol that leads to an artificially high pass rate. By revising the evaluation protocol to better reflect practical application scenarios, we introduce an additional benchmark, termed AnyToolBench. Experiments across various datasets demonstrate the superiority of our AnyTool over strong baselines such as ToolLLM and a GPT-4 variant tailored for tool utilization. For instance, AnyTool outperforms ToolLLM by +35.4% in terms of average pass rate on ToolBench. Code will be available at https://github.com/dyabel/AnyTool.