FinToolBench: Evaluating LLM Agents for Real-World Financial Tool Use

2026-03-09

AI Summary

This work addresses the challenge of evaluating large language model agents in finance, where existing methods fall short in assessing compliance and dynamic operational capabilities within realistic, executable tool environments. To bridge this gap, the authors introduce the first runnable benchmark grounded in real-world financial scenarios, integrating 760 executable financial APIs and 295 complex, tool-requiring queries. They propose a multidimensional evaluation framework that incorporates temporal relevance, intent categorization, and regulatory compliance. A key innovation is the Financial-Aware Tool Retrieval and Reasoning (FATR) mechanism, which enhances agent decision-making with domain-specific awareness. Additionally, the study releases the first open-source testing platform supporting auditable tool invocations, substantially improving the reliability of evaluations in terms of compliance, stability, and task completion quality.

Technology Category

• Natural Language Processing: Safety and Robustness
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Mechanism Design

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

The integration of Large Language Models (LLMs) into the financial domain is driving a paradigm shift from passive information retrieval to dynamic, agentic interaction. While general-purpose tool learning has witnessed a surge in benchmarks, the financial sector, characterized by high stakes, strict compliance, and rapid data volatility, remains critically underserved. Existing financial evaluations predominantly focus on static textual analysis or document-based QA, ignoring the complex reality of tool execution. Conversely, general tool benchmarks lack the domain-specific rigor required for finance, often relying on toy environments or a negligible number of financial APIs. To bridge this gap, we introduce FinToolBench, the first real-world, runnable benchmark dedicated to evaluating financial tool learning agents. Unlike prior works limited to a handful of mock tools, FinToolBench establishes a realistic ecosystem coupling 760 executable financial tools with 295 rigorous, tool-required queries. We propose a novel evaluation framework that goes beyond binary execution success, assessing agents on finance-critical dimensions: timeliness, intent type, and regulatory domain alignment. Furthermore, we present FATR, a finance-aware tool retrieval and reasoning baseline that enhances stability and compliance. By providing the first testbed for auditable, agentic financial execution, FinToolBench sets a new standard for trustworthy AI in finance. The tool manifest, execution environment, and evaluation code will be open-sourced to facilitate future research.

Problem

• financial tool use
• LLM agents
• evaluation benchmark
• real-world execution
• domain-specific rigor

Innovation

• FinToolBench
• financial tool learning
• LLM agents
• executable benchmark
• compliance-aware evaluation