MCP-Universe — leaderboard
Benchmark for LLMs and agents using real-world Model Context Protocol servers across location navigation, repository management, finance, 3D design, browser automation, and web search tasks.
Metric: Overall Success Rate (self-reported). Source: benchmarklist.com. Status: saturation imminent. 27 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 44.16 |
| 2 | Grok 4 | 33.33 |
| 3 | Claude Sonnet 4 (Thinking) | 30.3 |
| 4 | Claude Sonnet 4 | 29.44 |
| 5 | Claude Opus 4.1 | 29.44 |
| 6 | Grok 4 Fast | 27.27 |
| 7 | Grok Code Fast 1 | 26.41 |
| 8 | O3 (Medium) | 26.41 |
| 9 | GLM-4.6 | 25.97 |
| 10 | O4 Mini (Medium) | 25.97 |
| 11 | GLM-4.5 | 24.68 |
| 12 | Claude 3.7 Sonnet | 24.24 |
| 13 | Gemini 2.5 Pro | 22.08 |
| 14 | DeepSeek V3.1 | 22.08 |
| 15 | Gemini 2.5 Flash | 21.65 |
Interactive version: theaggregate.ai/benchmark?slug=mcp-universe · How the rankings work · Data refreshed daily, snapshot 2026-07-22.