MCPMark

MCPMark - GitHub MCP v0.30.3; Playwright responses truncated at 32K tokens.

Category

General Agents

Max Score

100.0

Score Type

percent

Active

Yes

Model Scores

Scores for MCPMark

Model Score Date Verified
Qwen3.6-35B-A3B
37.64%
16.04.2026 Verified
Gemma4-31B-it
17.14%
16.04.2026 Unverified
Qwen3.5-35B-A3B
26.79%
16.04.2026 Unverified
Gemma4-26B-A4B
12.91%
16.04.2026 Unverified
Qwen3.5-27B
36.88%
16.04.2026 Unverified
Kimi K3
100.00%
18.08.2026 Verified
Kimi K2.6
58.13%
23.08.2026 Verified
GPT-5.4
65.29%
23.08.2026 Verified
Claude Opus 4.6
59.00%
23.08.2026 Verified
Gemini 3.1 Pro
58.13%
23.08.2026 Verified
Kimi K2.5
29.50%
23.08.2026 Verified
Qwen3.5-9B
12.04%
— Verified
Spark-X2.5-4B
12.91%
— Verified
Spark-X2.5-1.7B
2.30
— Verified
Qwen3.5-4B
11.06%
— Verified