Skip to content

Benchmark

AndroidBench

Scores are percentages. Higher is better.

Models measured
5
Scores recorded
6
Sources
2
IQ weight
0.46

Score distribution

72.5%
Mean
75.1%
Median
8.4%
Std dev
69.8%
10th percentile
84.5%
90th percentile
56.5%
Min
56.5%84.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.46 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
1Claude Fable 5Qwen表:95题公开子集 · avg@3· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.84.5%
2Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.1%
3Qwen3.8 MaxQwen表:95题公开子集 · avg@3· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.75.1%
4GPT-5.6 SolQwen表:95题公开子集 · avg@3· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74%
5Claude Opus 4.8Qwen表:95题公开子集 · avg@3· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69.8%
6Qwen3.7 MaxQwen表:95题公开子集 · avg@3· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.56.5%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats