OSWorld 2.0 is an AI benchmark used to evaluate model capabilities. Review its overview, metrics, official resources, and model leaderboard results on DataLearnerAI.
Browse the latest scores, model modes, release dates, and parameter sizes for OSWorld 2.0.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
— |  Claude Fable 5.1 Thinking Level · MaxTools | 77.90 | 2026-09-01 | Unknown | Closed |
— |  GPT-6 Astra Thinking Level · MaxTools | 72.60 | 2026-09-03 | Unknown | Closed |
— | Claude Opus 5 Thinking Level · MaxTools | 70.57 | 2026-07-24 | Unknown | Closed |
— |  Muse Spark 1.3 Thinking Level · MaxTools | 66.90 | 2026-09-02 | Unknown | Closed |
— | GPT-5.6 Sol unknown | 65.70 | 2026-06-26 | Unknown | Closed |
— | GPT-5.6 Sol Thinking Level · Extra HighTools | 62.60 | 2026-06-26 | Unknown | Closed |
— |  Gemini 3.8 Flash Thinking EnabledTools | 59.00 | 2026-09-02 | Unknown | Closed |
— |  Kimi K3 Thinking Level · MaxTools | 58.30 | 2026-07-16 | 2800B | Conditional |
— |  Qwen3.8-Flash-Next Thinking Level · Extra HighTools | 52.30 | 2026-08-26 | 125B | Conditional |
— | Gemini 3.7 Flash Thinking Level · MediumTools | 47.90 | 2026-08-13 | Unknown | Closed |
