Skip to content

Benchmark

EvasionBench

Scores are percentages. Higher is better.

Models measured
5
Scores recorded
5
Sources
2
IQ weight
0.11

Score distribution

73.9%
Mean
71.3%
Median
6.4%
Std dev
66.7%
10th percentile
82.9%
90th percentile
66.7%
Min
66.7%82.9%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.11 for chat models, measured over 5 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GLM 4.7· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
82.9%
2Qwen3 Coder· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
78.2%
3MiniMax M2.1· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
71.3%
4DeepSeek V3.2· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it.66.9%
5Kimi K2 Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
66.7%

Data credits: OpenEvals/leaderboard-data, MIT (licence) · Hugging Face official benchmark leaderboards; licence varies by benchmark