Help or Hurdle? Rethinking Model Context Protocol-Augmented Large Language Models
This work addresses the need for better evaluation of tool-augmented LLMs for researchers and developers, though it is incremental as it focuses on benchmarking rather than proposing new methods.
The paper tackles the problem of understanding how large language models (LLMs) leverage the Model Context Protocol (MCP) for tool integration, revealing critical limitations in current AI-tool integration through a large-scale evaluation with 20,000 API calls and $6,000 in cost.
The Model Context Protocol (MCP) enables large language models (LLMs) to access external resources on demand. While commonly assumed to enhance performance, how LLMs actually leverage this capability remains poorly understood. We introduce MCPGAUGE, the first comprehensive evaluation framework for probing LLM-MCP interactions along four key dimensions: proactivity (self-initiated tool use), compliance (adherence to tool-use instructions), effectiveness (task performance post-integration), and overhead (computational cost incurred). MCPGAUGE comprises a 160-prompt suite and 25 datasets spanning knowledge comprehension, general reasoning, and code generation. Our large-scale evaluation, spanning six commercial LLMs, 30 MCP tool suites, and both one- and two-turn interaction settings, comprises around 20,000 API calls and over USD 6,000 in computational cost. This comprehensive study reveals four key findings that challenge prevailing assumptions about the effectiveness of MCP integration. These insights highlight critical limitations in current AI-tool integration and position MCPGAUGE as a principled benchmark for advancing controllable, tool-augmented LLMs.