# OSWorld-Verified Leaderboard

> Last updated AUG 07, 2026

A current ranking of officially verified computer-use results. Intelligence-Indeed Agent leads the full board at 90.2%.

Data is reproduced from the official OSWorld-Verified workbook maintained by the OSWorld team. Yutori Scouts monitor the source for changes.

## Foundation E2E GUI

General and specialized models without extra accessibility-tree or coding-action channels.

| Model | Institution | Approach type | Max steps | Success rate | Std. dev. | Runs | Date | A11y tree | Coding action | Multi-rollout |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [claude-fable-5\[1m\]](<https://www.anthropic.com/claude/fable>) | Anthropic | General model | 100 | 86.0% | 0.0 | 1 | Aug 01, 2026 | No | No | No |
| [claude-opus-5\[1m\]](<https://www.anthropic.com/news/claude-opus-5>) | Anthropic | General model | 100 | 83.4% | 0.0 | 1 | Aug 01, 2026 | No | No | No |
| [Muse Spark 1.1](<https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/>) | Meta Superintelligence Labs | General model | 100 | 80.7% | 0.0 | 1 | Jul 09, 2026 | No | No | No |
| [Holo3-35B-A3B](<https://hcompany.ai/holo3>) | H Company | Specialized model | 100 | 80.4% | 2.2 | 2 | Apr 20, 2026 | No | No | No |
| [MiniMax M3](<https://www.minimaxi.com/blog/minimax-m3>) | MiniMax | General model | 100 | 75.2% | 0.0 | 1 | Jun 08, 2026 | No | No | No |
| [Qwen 3.7 Plus](<https://qwen.ai/blog?id=qwen3.7>) | Qwen Team, Alibaba Group | General model | 100 | 73.3% | 0.0 | 1 | May 25, 2026 | No | No | No |
| [Kimi K2.6](<https://www.kimi.com/blog/kimi-k2-6>) | Moonshot AI | General model | 100 | 73.1% | 0.0 | 1 | Apr 20, 2026 | No | No | No |
| [claude-sonnet-4-6](<https://www.anthropic.com/news/claude-sonnet-4-6>) | Anthropic | General model | 100 | 72.1% | 0.0 | 1 | Mar 08, 2026 | No | No | No |
| [Kimi K2.5](<https://github.com/MoonshotAI/Kimi-K2.5/blob/master/tech_report.pdf>) | Moonshot AI | General model | 100 | 63.3% | 0.0 | 1 | Jan 31, 2026 | No | No | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 100 | 62.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [Seed-1.8](<https://seed.bytedance.com/en/seed1_8>) | ByteDance Seed | General model | 100 | 61.9% | 0.0 | 1 | Dec 18, 2025 | No | No | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 50 | 58.1% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [EvoCUA-20260105](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 56.7% | 0.0 | 1 | Jan 06, 2026 | No | No | No |
| [GUI-Owl-1.5 32B](<https://arxiv.org/abs/2602.16855>) | Tongyi Lab, Alibaba Group | Specialized model | 50 | 55.4% | 0.0 | 1 | Mar 27, 2026 | No | No | No |
| [DeepMiner-Mano-72B](<https://arxiv.org/abs/2509.17336>) | Mininglamp Technology | Specialized model | 100 | 53.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [UI-TARS-2-2509](<https://arxiv.org/pdf/2509.02544>) | ByteDance Seed | Specialized model | 100 | 53.1% | 0.0 | 1 | Oct 14, 2025 | No | No | No |
| [EvoCUA](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 50.3% | 0.0 | 1 | Dec 24, 2025 | No | No | No |
| [EvoCUA-8B-20260105](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 46.1% | 0.0 | 1 | Jan 13, 2026 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 45.0% | 0.9 | 3 | Oct 01, 2025 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 44.9% | 0.0 | 1 | Oct 01, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 50 | 43.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 15 | 42.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [UI-TARS-250705](<https://arxiv.org/pdf/2509.02544>) | ByteDance Seed | General model | 100 | 41.8% | 0.0 | 1 | Sep 16, 2025 | No | No | No |
| [qwen3-vl-flash-2025-10-25](<https://qwen.ai/blog?id=99f0335c4ad9ff6153e517418d48535ab6d8afef&from=research.latest-advancements-list>) | Qwen Team, Alibaba Group | General model | 100 | 41.6% | 0.0 | 1 | Oct 23, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 100 | 41.4% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [DART-GUI-7B-0924](<https://arxiv.org/pdf/2509.23866>) | BIGAI & BIT & DataCanvas | Specialized model | 30 | 40.5% | 0.0 | 1 | Jan 23, 2026 | No | No | No |
| [DeepMiner-Mano-7B](<https://arxiv.org/abs/2509.17336>) | Mininglamp Technology | Specialized model | 100 | 40.1% | 0.0 | 1 | Sep 18, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250717](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 100 | 40.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 39.0% | 0.0 | 1 | Oct 01, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 50 | 35.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 100 | 35.6% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 34.8% | 0.8 | 3 | Jul 28, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 34.1% | 0.7 | 3 | Jul 28, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250428](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 100 | 33.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [gui-owl-7b](<https://arxiv.org/abs/2508.15144>) | Tongyi Lab, Alibaba Group | Specialized model | 15 | 32.1% | 0.0 | 1 | Sep 03, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250717](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 15 | 31.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 50 | 31.3% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 15 | 31.2% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 100 | 30.5% | 0.9 | 2 | Jul 28, 2025 | No | No | No |
| [TianXi-Action-7B](<http://arxiv.org/abs/2508.04037>) | Lenovo | Specialized model | 50 | 29.8% | 0.6 | 2 | Aug 11, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 29.7% | 1.1 | 3 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 28.2% | 0.5 | 3 | Jul 28, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250428](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 15 | 27.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 100 | 27.4% | 2.2 | 2 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 50 | 27.3% | 2.1 | 2 | Jul 28, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 15 | 27.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 100 | 27.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 26.6% | 0.4 | 3 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 15 | 26.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 50 | 25.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 15 | 24.5% | 1.2 | 2 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 24.3% | 1.4 | 3 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 15 | 24.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 23.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 100 | 23.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 20.6% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 19.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 19.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 17.7% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 50 | 17.2% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 16.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [kimi-vl-a3b](<https://arxiv.org/abs/2504.07491>) | Moonshot AI | General model | 100 | 10.3% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [kimi-vl-a3b](<https://arxiv.org/abs/2504.07491>) | Moonshot AI | General model | 15 | 9.7% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 15 | 9.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-72b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 100 | 5.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-72b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 15 | 4.4% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-32b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 100 | 3.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-32b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 15 | 3.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |

## All verified results

Every numeric result in the official verified workbook, including agentic and augmented approaches.

| Model | Institution | Approach type | Max steps | Success rate | Std. dev. | Runs | Date | A11y tree | Coding action | Multi-rollout |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| [Intelligence-Indeed Agent](<https://github.com/intelligence-indeed/intelligence-indeed>) | Intelligence Indeed | Agentic framework | 100 | 90.2% | 0.0 | 1 | Jul 25, 2026 | No | Yes | No |
| [claude-fable-5\[1m\]](<https://www.anthropic.com/claude/fable>) | Anthropic | General model | 100 | 86.0% | 0.0 | 1 | Aug 01, 2026 | No | No | No |
| [Pointer Agent w/ Opus 4.7](<https://pointer.ai/blog/sota?utm_source=osworld&utm_medium=leaderboard&utm_campaign=sota>) | Pointer | Agentic framework | 100 | 83.6% | 0.0 | 1 | May 21, 2026 | No | Yes | No |
| [claude-opus-5\[1m\]](<https://www.anthropic.com/news/claude-opus-5>) | Anthropic | General model | 100 | 83.4% | 0.0 | 1 | Aug 01, 2026 | No | No | No |
| [Coasty CUA v1](<https://coasty.ai>) | Coasty Team | General model | 100 | 82.8% | 0.0 | 1 | Jul 01, 2026 | No | Yes | No |
| [Pointer Agent w/ Sonnet 4.6](<https://pointer.ai/blog/sota?utm_source=osworld&utm_medium=leaderboard&utm_campaign=sota>) | Pointer | Agentic framework | 100 | 81.5% | 0.0 | 1 | May 21, 2026 | No | Yes | No |
| [Muse Spark 1.1](<https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/>) | Meta Superintelligence Labs | General model | 100 | 80.7% | 0.0 | 1 | Jul 09, 2026 | No | No | No |
| [Holo3-35B-A3B](<https://hcompany.ai/holo3>) | H Company | Specialized model | 100 | 80.4% | 2.2 | 2 | Apr 20, 2026 | No | No | No |
| [OpenAPA w/ gemini-3.1-pro](<https://github.com/laiye-ai/open-apa>) | Laiye | Agentic framework | 100 | 78.3% | 0.0 | 1 | Apr 17, 2026 | No | Yes | No |
| [VLAA-GUI w/ Opus 4.5](<https://ucsc-vlaa.github.io/VLAA-GUI/>) | UCSC VLAA Lab | Agentic framework | 100 | 76.3% | 0.0 | 1 | Apr 20, 2026 | No | Yes | No |
| [MiniMax M3](<https://www.minimaxi.com/blog/minimax-m3>) | MiniMax | General model | 100 | 75.2% | 0.0 | 1 | Jun 08, 2026 | No | No | No |
| [HIPPO Agent w/ Opus 4.5](<https://github.com/wadang/muscle-mem-agent>) | Lenovo | Agentic framework | 100 | 74.5% | 0.0 | 1 | Feb 25, 2026 | No | Yes | No |
| [Qwen 3.7 Plus](<https://qwen.ai/blog?id=qwen3.7>) | Qwen Team, Alibaba Group | General model | 100 | 73.3% | 0.0 | 1 | May 25, 2026 | No | No | No |
| [Kimi K2.6](<https://www.kimi.com/blog/kimi-k2-6>) | Moonshot AI | General model | 100 | 73.1% | 0.0 | 1 | Apr 20, 2026 | No | No | No |
| [agent s3 w/ Opus 4.5 + GPT-5 bBoN (N=10)](<https://arxiv.org/abs/2510.02250>) | Simular | Agentic framework | 100 | 72.6% | 0.0 | 1 | Dec 11, 2025 | No | Yes | Yes |
| [claude-sonnet-4-6](<https://www.anthropic.com/news/claude-sonnet-4-6>) | Anthropic | General model | 100 | 72.1% | 0.0 | 1 | Mar 08, 2026 | No | No | No |
| [agent s3 w/ GPT-5 bBoN (N=10)](<https://arxiv.org/abs/2510.02250>) | Simular | Agentic framework | 100 | 69.9% | 0.0 | 1 | Oct 04, 2025 | No | Yes | Yes |
| [UiPath Screen Agent w/ Opus 4.5](<https://github.com/xlang-ai/OSWorld/tree/main/mm_agents/uipath>) | UiPath | Agentic framework | 100 | 67.1% | 0.0 | 1 | Dec 24, 2025 | No | No | No |
| [agent s3 w/ Opus 4.5 bBoN (N=1)](<https://arxiv.org/abs/2510.02250>) | Simular | Agentic framework | 100 | 66.0% | 1.1 | 5 | Dec 11, 2025 | No | Yes | No |
| [OS-Symphony w/ GPT-5](<https://github.com/xlang-ai/OSWorld/tree/main/mm_agents/os_symphony>) | Shanghai AI Laboratory | Agentic framework | 50 | 65.8% | 0.0 | 1 | Jan 04, 2026 | No | Yes | No |
| [UiPath Screen Agent w/ Opus 4.5](<https://github.com/xlang-ai/OSWorld/tree/main/mm_agents/uipath>) | UiPath | Agentic framework | 50 | 64.4% | 0.0 | 1 | Dec 24, 2025 | No | No | No |
| [GBOX Agent](<https://github.com/babelcloud/GBOXBlog/blob/main/From_Vision_to_Action.md>) | GBOX.AI | Agentic framework | 15 | 64.2% | 0.0 | 1 | Nov 25, 2025 | No | No | No |
| [GTA1 w/ GPT-5](<https://arxiv.org/pdf/2507.05791>) | Salesforce & The Australian National University & The University of Hong Kong | Agentic framework | 100 | 63.4% | 0.0 | 1 | Oct 03, 2025 | No | Yes | No |
| [Kimi K2.5](<https://github.com/MoonshotAI/Kimi-K2.5/blob/master/tech_report.pdf>) | Moonshot AI | General model | 100 | 63.3% | 0.0 | 1 | Jan 31, 2026 | No | No | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 100 | 62.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [agent s3 w/ GPT-5 bBoN (N=1)](<https://arxiv.org/abs/2510.02250>) | Simular | Agentic framework | 100 | 62.6% | 1.5 | 10 | Oct 04, 2025 | No | Yes | No |
| [Agentic-Lybic-Maestro](<https://arxiv.org/abs/2509.11067>) | Lybic | Agentic framework | 100 | 61.9% | 0.0 | 1 | Oct 17, 2025 | No | Yes | No |
| [Seed-1.8](<https://seed.bytedance.com/en/seed1_8>) | ByteDance Seed | General model | 100 | 61.9% | 0.0 | 1 | Dec 18, 2025 | No | No | No |
| [CoACT-1](<https://arxiv.org/abs/2508.03923>) | Salesforce & University of Southern California & University of Washington | Agentic framework | 150 | 60.8% | 0.0 | 1 | Aug 04, 2025 | No | Yes | No |
| [CoACT-1](<https://arxiv.org/abs/2508.03923>) | Salesforce & University of Southern California & University of Washington | Agentic framework | 100 | 59.9% | 0.0 | 1 | Aug 04, 2025 | No | Yes | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 50 | 58.1% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [aworldGUIAgent-v1](<https://github.com/inclusionAI/AWorld/tree/main/examples/osworld>) | inclusionAI | Agentic framework | 50 | 58.0% | 0.0 | 1 | Sep 18, 2025 | No | Yes | No |
| [Agentic-Lybic-Maestro](<https://arxiv.org/abs/2509.11067>) | Lybic | Agentic framework | 50 | 56.9% | 0.2 | 2 | Oct 17, 2025 | No | Yes | No |
| [EvoCUA-20260105](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 56.7% | 0.0 | 1 | Jan 06, 2026 | No | No | No |
| [CoACT-1](<https://arxiv.org/abs/2508.03923>) | Salesforce & University of Southern California & University of Washington | Agentic framework | 50 | 56.4% | 0.0 | 1 | Aug 04, 2025 | No | Yes | No |
| [agent s2.5 w/ o3](<https://github.com/simular-ai/Agent-S>) | Simular Research | Agentic framework | 100 | 56.0% | 0.0 | 1 | Jul 31, 2025 | No | No | No |
| [GUI-Owl-1.5 32B](<https://arxiv.org/abs/2602.16855>) | Tongyi Lab, Alibaba Group | Specialized model | 50 | 55.4% | 0.0 | 1 | Mar 27, 2026 | No | No | No |
| [agent s2.5 w/ o3](<https://github.com/simular-ai/Agent-S>) | Simular Research | Agentic framework | 50 | 54.2% | 0.0 | 1 | Jul 31, 2025 | No | No | No |
| [DeepMiner-Mano-72B](<https://arxiv.org/abs/2509.17336>) | Mininglamp Technology | Specialized model | 100 | 53.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| UiPath Screen Agent w/ GPT-5 | UiPath | Agentic framework | 50 | 53.6% | 0.0 | 1 | Sep 14, 2025 | No | No | No |
| [GTA1 w/ o3](<https://arxiv.org/abs/2507.05791>) | Salesforce & The Australian National University & The University of Hong Kong | Agentic framework | 100 | 53.1% | 0.0 | 1 | Jul 28, 2025 | No | Yes | No |
| [UI-TARS-2-2509](<https://arxiv.org/pdf/2509.02544>) | ByteDance Seed | Specialized model | 100 | 53.1% | 0.0 | 1 | Oct 14, 2025 | No | No | No |
| [Jedi-7B w/ o3](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 100 | 51.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [Jedi-7B w/ o3](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 50 | 50.6% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [EvoCUA](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 50.3% | 0.0 | 1 | Dec 24, 2025 | No | No | No |
| [GTA1 w/ o3](<https://arxiv.org/abs/2507.05791>) | Salesforce & The Australian National University & The University of Hong Kong | Agentic framework | 50 | 48.6% | 0.0 | 1 | Jul 28, 2025 | No | Yes | No |
| [autoglm-os-9b-20250925](<https://arxiv.org/pdf/2508.14040>) | Tsinghua University & Zhipu AI | Specialized model | 50 | 48.0% | 0.0 | 1 | Sep 25, 2025 | No | Yes | No |
| [autoglm-os-9b](<https://arxiv.org/pdf/2508.14040>) | Tsinghua University & Zhipu AI | Specialized model | 50 | 47.3% | 0.0 | 1 | Aug 18, 2025 | Yes | Yes | No |
| [autoglm-os-9b-20250925](<https://arxiv.org/pdf/2508.14040>) | Tsinghua University & Zhipu AI | Specialized model | 15 | 46.9% | 0.0 | 1 | Sep 25, 2025 | No | Yes | No |
| [autoglm-os-9b](<https://arxiv.org/pdf/2508.14040>) | Tsinghua University & Zhipu AI | Specialized model | 15 | 46.3% | 0.0 | 1 | Aug 18, 2025 | Yes | Yes | No |
| [EvoCUA-8B-20260105](<https://arxiv.org/abs/2601.15876>) | Meituan LongCat Team | General model | 50 | 46.1% | 0.0 | 1 | Jan 13, 2026 | No | No | No |
| [agent s2 w/ gemini-2.5-pro](<https://github.com/simular-ai/Agent-S>) | Simular Research | Agentic framework | 50 | 45.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 45.0% | 0.9 | 3 | Oct 01, 2025 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 44.9% | 0.0 | 1 | Oct 01, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 50 | 43.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-sonnet-4-5-20250929](<https://www.anthropic.com/news/claude-sonnet-4-5>) | Anthropic | General model | 15 | 42.9% | 0.0 | 1 | Oct 31, 2025 | No | No | No |
| [Jedi-7B w/ o3](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 15 | 42.4% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [UI-TARS-250705](<https://arxiv.org/pdf/2509.02544>) | ByteDance Seed | General model | 100 | 41.8% | 0.0 | 1 | Sep 16, 2025 | No | No | No |
| [qwen3-vl-flash-2025-10-25](<https://qwen.ai/blog?id=99f0335c4ad9ff6153e517418d48535ab6d8afef&from=research.latest-advancements-list>) | Qwen Team, Alibaba Group | General model | 100 | 41.6% | 0.0 | 1 | Oct 23, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 100 | 41.4% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [DART-GUI-7B-0924](<https://arxiv.org/pdf/2509.23866>) | BIGAI & BIT & DataCanvas | Specialized model | 30 | 40.5% | 0.0 | 1 | Jan 23, 2026 | No | No | No |
| [DeepMiner-Mano-7B](<https://arxiv.org/abs/2509.17336>) | Mininglamp Technology | Specialized model | 100 | 40.1% | 0.0 | 1 | Sep 18, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250717](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 100 | 40.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [CoACT-1](<https://arxiv.org/abs/2508.03923>) | Salesforce & University of Southern California & University of Washington | Agentic framework | 15 | 39.8% | 0.0 | 1 | Aug 04, 2025 | No | Yes | No |
| [agent s2.5 w/ o3](<https://github.com/simular-ai/Agent-S>) | Simular Research | Agentic framework | 15 | 39.0% | 0.0 | 1 | Jul 31, 2025 | No | No | No |
| [opencua-72b-preview](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 39.0% | 0.0 | 1 | Oct 01, 2025 | No | No | No |
| [mobile-agent-v3 w/ gui-owl-32b](<https://arxiv.org/abs/2508.15144>) | Tongyi Lab, Alibaba Group | Agentic framework | 50 | 38.9% | 0.0 | 1 | Sep 03, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 50 | 35.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 100 | 35.6% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 34.8% | 0.8 | 3 | Jul 28, 2025 | No | No | No |
| [agent s2 w/ gemini-2.5-pro](<https://github.com/simular-ai/Agent-S>) | Simular Research | Agentic framework | 15 | 34.6% | 0.0 | 1 | Jul 31, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 34.1% | 0.7 | 3 | Jul 28, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250428](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 100 | 33.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [gui-owl-7b](<https://arxiv.org/abs/2508.15144>) | Tongyi Lab, Alibaba Group | Specialized model | 15 | 32.1% | 0.0 | 1 | Sep 03, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250717](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 15 | 31.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 50 | 31.3% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [claude-4-sonnet-20250514](<https://www.anthropic.com/news/claude-4>) | Anthropic | General model | 15 | 31.2% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 100 | 30.5% | 0.9 | 2 | Jul 28, 2025 | No | No | No |
| [TianXi-Action-7B](<http://arxiv.org/abs/2508.04037>) | Lenovo | Specialized model | 50 | 29.8% | 0.6 | 2 | Aug 11, 2025 | No | No | No |
| [opencua-32b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 29.7% | 1.1 | 3 | Jul 28, 2025 | No | No | No |
| [Jedi-7B w/ gpt-4o](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 100 | 29.3% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 28.2% | 0.5 | 3 | Jul 28, 2025 | No | No | No |
| [doubao-1-5-thinking-vision-pro-250428](<https://arxiv.org/abs/2505.07062>) | ByteDance Seed | General model | 15 | 27.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 100 | 27.4% | 2.2 | 2 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 50 | 27.3% | 2.1 | 2 | Jul 28, 2025 | No | No | No |
| [claude-3-7-sonnet-20250219](<https://www.anthropic.com/news/claude-3-7-sonnet>) | Anthropic | General model | 15 | 27.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 100 | 27.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [Jedi-7B w/ gpt-4o](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 50 | 27.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [Jedi-7B w/ gpt-4o](<https://osworld-grounding.github.io/>) | Salesforce & The University of Hong Kong | Agentic framework | 15 | 26.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 26.6% | 0.4 | 3 | Jul 28, 2025 | No | No | No |
| [computer-use-preview](<https://openai.com/index/computer-using-agent/>) | OpenAI | Specialized model | 15 | 26.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 50 | 25.8% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [uitars-1.5-7b](<https://seed-tars.com/1.5>) | ByteDance Seed & Tsinghua University | Specialized model | 15 | 24.5% | 1.2 | 2 | Jul 28, 2025 | No | No | No |
| [opencua-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 24.3% | 1.4 | 3 | Jul 28, 2025 | No | No | No |
| [uitars-72b-dpo](<https://github.com/bytedance/UI-TARS>) | ByteDance Seed & Tsinghua University | Specialized model | 15 | 24.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 23.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 100 | 23.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 20.6% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 50 | 19.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-qwen2-7b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 19.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 100 | 17.7% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 50 | 17.2% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [opencua-a3b](<https://opencua.xlang.ai/>) | The University of Hong Kong & Moonshot AI | Specialized model | 15 | 16.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [kimi-vl-a3b](<https://arxiv.org/abs/2504.07491>) | Moonshot AI | General model | 100 | 10.3% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [kimi-vl-a3b](<https://arxiv.org/abs/2504.07491>) | Moonshot AI | General model | 15 | 9.7% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [o3](<https://openai.com/index/introducing-o3-and-o4-mini/>) | OpenAI | General model | 15 | 9.1% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-72b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 100 | 5.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-72b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 15 | 4.4% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-32b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 100 | 3.9% | 0.0 | 1 | Jul 28, 2025 | No | No | No |
| [qwen2.5-vl-32b-instruct](<https://qwenlm.github.io/blog/qwen2.5-vl/>) | Alibaba Cloud, Qwen Team | General model | 15 | 3.0% | 0.0 | 1 | Jul 28, 2025 | No | No | No |

## About the benchmark

OSWorld-Verified is the corrected and centrally evaluated release of the original OSWorld desktop benchmark. It contains 369 tasks; the maintainers also permit a 361-task setup that excludes eight Google Drive tasks requiring manual configuration. The official team publishes verified trajectories for community analysis.

## Official sources

- [Paper (arXiv 2404.07972)](<https://arxiv.org/abs/2404.07972>)
- [Code (GitHub · XLANG)](<https://github.com/xlang-ai/OSWorld>)
- [Verified trajectories (Hugging Face)](<https://huggingface.co/datasets/xlangai/ubuntu_osworld_verified_trajs>)
- [Official leaderboard](<https://osworld-v1.xlang.ai/#benchmark>)

Source revision (SHA-256): `cf6b4b67eed566ddcd5a8b7ad2d89013157978dad0eea76ea12d2377471efc09`