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: saturated. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 (High) | 44.16 |
| 2 | GPT-5 (Medium) | 43.72 |
| 3 | Grok 4 | 33.33 |
| 4 | Claude Sonnet 4 (Thinking) | 30.3 |
| 5 | Claude Sonnet 4 | 29.44 |
| 6 | Claude Opus 4.1 | 29.44 |
| 7 | Grok 4 Fast | 27.27 |
| 8 | Grok Code Fast 1 | 26.41 |
| 9 | O3 (Medium) | 26.41 |
| 10 | GLM-4.6 | 25.97 |
| 11 | O4 Mini (Medium) | 25.97 |
| 12 | GLM-4.5 | 24.68 |
| 13 | Claude 3.7 Sonnet | 24.24 |
| 14 | Qwen 3 Coder 480B A35B Instruct | 22.94 |
| 15 | Gemini 2.5 Pro | 22.08 |
Interactive version: theaggregate.ai/benchmark?slug=mcp-universe · How It Works · Data refreshed daily, snapshot 2026-09-05.