MCP-Mark

agents

MCP-Mark evaluates LLMs on their ability to use Model Context Protocol (MCP) tools effectively, testing tool discovery, selection, invocation, and result interpretation across diverse MCP server scenarios.

Leaderboard

Showing 7 of 7 results

Kimi K2.7 Code

81.1%

i
Qwen3.7 Max

60.8%

i
Kimi K2.6

55.9%

i
Qwen3.6 Plus

48.2%

i
Qwen3.5-397B-A17B

46.1%

i
DeepSeek-V3.2

38.0%

i
Qwen3.6-35B-A3B

37.0%

i