Skip to content

Benchmark

AndroidWorld_SR

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

Models measured
7
Scores recorded
7
Sources
1
IQ weight
2.50

Score distribution

55.7%
Mean
63.7%
Median
16.4%
Std dev
35%
10th percentile
66.4%
90th percentile
25.5%
Min
25.5%71.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.50 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
1Qwen3.5 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.1%
2Qwen3.5 122B-A10B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim66.4%
3Qwen3.5 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.2%
4Qwen3 VL 235B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.63.7%
5Qwen3 VL 32B Thinking· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.63.7%
6Qwen2.5 VL 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim35%
7Qwen2.5-VL 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim25.5%

Data credits: Data by LLM Stats