MCP-Universe
Benchmark and framework that tests models and agents on tasks using real-world MCP servers.
Reference by Salesforce AI Research
IndexedApache-2.0★ 602
About
Whether a model or agent can finish tasks through live MCP servers in areas such as navigation, repositories, finance, 3D design, browsing and search.
Use cases: llm-evaluation
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
About the benchmark
- Benchmark site
- mcp-universe.github.io/
- Maintainer
- Salesforce AI Researchmcp-universe.github.io
- Launched
- 2025-08arxiv.org
- Leaderboard
- mcp-universe.github.io/
We link to the official leaderboard instead of copying scores. Benchmarks get gamed and saturate; we never use them to rank listings.
Signals
- Stars
- 602
- Forks
- 88
- Open issues
- 18
- Last commit
- Oct 8, 2026 (20 days ago)
- Last release
- Mar 25, 2026 (7 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Website
- mcp-universe.github.io/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
Cite this page
MCP-Universe benchmark. Agent Mart. https://agentmart.dev/reference/mcp-universe (last updated 2026-10-24)