Skip to content

Benchmark

OSWorld 2.1 (partial)

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

Models measured
8
Scores recorded
8
Sources
1
IQ weight
1.29

Score distribution

76.9%
Mean
78.4%
Median
6.3%
Std dev
72.4%
10th percentile
81.5%
90th percentile
68.1%
Min
68.1%89.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.29 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

#ModelScore
1Claude Fable 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim89.3%
2Gemini 4 Carbon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.5%
3Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.1%
4Muse Wonder Full· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.4%
5Muse Wonder· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.6%
6Muse Wonder 1.0· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.6%
7Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.4%
8Hy4 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68.1%

Data credits: Data by LLM Stats