Skip to content

Benchmark

MCPMark Verified GitHub Pass@1

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
8
Scores recorded
8
Sources
1
IQ weight
0.54

Score distribution

76.1%
Mean
82.6%
Median
18.1%
Std dev
56.5%
10th percentile
95.7%
90th percentile
47.8%
Min
47.8%95.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.54 for chat models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model
1GPT 5.5config: gpt-5-5-xhigh, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs95.7%
2GPT-5.6 Solconfig: gpt-5-6-sol-max, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs95.7%
3Kimi K3config: kimi-k3-max, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs95.7%
4Claude Fable 5config: claude-fable-5-max, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs82.6%
5Kimi K2.7 Codeconfig: kimi-k2-7-code, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs73.9%
6Kimi K2.6config: kimi-k2-6, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs60.9%
7Claude Opus 4.8config: claude-opus-4-8-max, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs56.5%
8DeepSeek V4 Proconfig: deepseek-v4-pro-max, task-set: standard, task-version: verified, scope: github· MCPMark Verified leaderboard docs47.8%

Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.

Data credits: MCPMark (eval-sys), mcpmark.ai Verified leaderboard. Single-run Pass@1 fractions shown as percentages.