MCP-Universe
A benchmark from Salesforce AI Research that evaluates LLMs and agents on real-world Model Context Protocol (MCP) server tasks across six domains (location navigation, repository management, financial analysis, 3D design, browser automation, web searching), measuring end-to-end task success rate.
What this benchmark measures
A benchmark from Salesforce AI Research that evaluates LLMs and agents on real-world Model Context Protocol (MCP) server tasks across six domains (location navigation, repository management, financial analysis, 3D design, browser automation, web searching), measuring end-to-end task success rate.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is Overall Success Rate. It should be interpreted within MCP-Universe, not compared as part of a site-wide ranking.
Frequently asked
What is MCP-Universe?
A benchmark from Salesforce AI Research that evaluates LLMs and agents on real-world Model Context Protocol (MCP) server tasks across six domains (location navigation, repository management, financial analysis, 3D design, browser automation, web searching), measuring end-to-end task success rate. It is a tool use benchmark measured by Overall Success Rate.
What does Overall Success Rate mean on MCP-Universe?
MCP-Universe reports Overall Success Rate (%); higher is better. Scores are shown only within MCP-Universe and are never averaged with other benchmarks.
What is the top reported MCP-Universe score?
Gemini 3 Pro has the top reported score on MCP-Universe: 44.59% (Overall Success Rate).
Why do MCP-Universe scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. MCP-Universe scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".