Skip to content

Benchmark

OSWorld 2.0

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

Models measured
21
Scores recorded
28
Sources
2
IQ weight
1.81

Score distribution

58.2%
Mean
61.7%
Median
14.3%
Std dev
45.6%
10th percentile
71.4%
90th percentile
19.4%
Min
19.4%81.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.81 for chat models, measured over 18 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 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.8%
2Claude Fable 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.9%
3GPT-6 Astra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim72.6%
4GPT-6.1 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.4%
5Claude Opus 5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.6%
6Claude Opus 5Opus 5 System Card setting· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70.6%
7Gemini 4 Argon· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.2%
8Muse Spark 1.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim66.9%
9Claude Fable 5Opus 5 System Card setting· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.66.1%
10GPT-6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.4%
11GPT-5.6 Sol· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.6%
12GPT-5.6 SolOpus 5 System Card setting· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.6%
13GPT-5.6 SolOSWorld 2.0 · OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.6%
14MiniMax M3.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim61.7%
15MiniMax M3.1 FlashMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · OSWorld 2.0 0808 Full release · highest reported at release · metric subtype unstated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.7%
16Gemini 3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59%
17Kimi K3max effort · temp 1.0 · top_p 1.0· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.58.3%
18Claude Opus 4.8Opus 5 System Card setting· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55.7%
19Claude Opus 4.8OSWorld 2.0 · OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.54.8%
20GPT-6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim52.7%
21GPT-5.6 Terra· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim50.2%
22GPT-5.6 TerraOSWorld 2.0 · OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.2%
23Gemini 3.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim47.9%
24GPT 5.5OSWorld 2.0 · OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.47.5%
25GPT-5.6 Luna· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim45.6%
26GPT-5.6 LunaOSWorld 2.0 · OpenAI GPT-5.6 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.6%
27Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim19.4%
28Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim19.4%

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