Skip to content

Benchmark

Graphwalks parents >128k

Scores are percentages.

Models measured
7
Scores recorded
7
Sources
1
IQ weight
0.85

Score distribution

44.5%
Mean
32.4%
Median
32.6%
Std dev
11%
10th percentile
83.3%
90th percentile
5.6%
Min
5.6%95.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.85 for chat models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Opus 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.4%
2Claude Opus 4.8· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.3%
3GPT 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim58.5%
4GPT 5.4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim32.4%
5GPT 4.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim25%
6GPT-4.1 mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim11%
7GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim5.6%

Data credits: Data by LLM Stats