A recent analysis of 34,374 tool lists from MCP servers revealed significant cost implications due to token usage. While the median server uses only 8 tools, the mean is 18.5, with a few servers listing hundreds of tools and definitions that can consume thousands of tokens each. This extensive token usage, which rides along with every model request, can lead to substantial costs, especially for clients with multiple servers. Strategies like prompt caching, on-demand definition loading, and optimizing definition size can help mitigate these expenses. AI
IMPACT High token usage in tool lists can increase operational costs for AI applications, prompting developers to optimize definitions and caching strategies.
RANK_REASON Analysis of token costs for MCP server tool lists. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →