OSWorld
Updated
Cite
Fill in the access date before citing.
Plain text
Benchmark Registry. OSWorld 2.0 v2026.08.08 Offline — Partial benchmark results. Benchmark identifier: osworld/2.0-v2026.08.08-offline-partial. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes.
BibTeX
@misc{br_osworld_2_0_v2026_08_08_offline_partial,
title = {OSWorld 2.0 v2026.08.08 Offline — Partial benchmark results},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial},
urldate = {YYYY-MM-DD},
note = {Benchmark identifier: osworld/2.0-v2026.08.08-offline-partial; data license: CC BY 4.0; access date to be filled by the reader}
}Results
31 results
| Provider | Model | Score | Source | Registry No. |
|---|---|---|---|---|
| OpenAI | GPT-6.1 Sol (high) | 69.6% | openai.com for GPT-6.1 Sol (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.6% (high). Registry No. 10015; Record No. 45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638#BR-45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638,
title = {GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.6\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638\#BR-45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638},
urldate = {YYYY-MM-DD},
note = {Registry No. 10015; Record No. 45a86a2c6cfbac0c6115981fa239d55a1ac95e264606a725a29e70e9ce63c638; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10015/osworld.svg" alt="GPT-6.1 Sol: OSWorld" height="24"> | 10015 |
| OpenAI | GPT-6.1 Sol (xhigh) | 69.4% | openai.com for GPT-6.1 Sol (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.4% (xhigh). Registry No. 10015; Record No. 56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7#BR-56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7,
title = {GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.4\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7\#BR-56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7},
urldate = {YYYY-MM-DD},
note = {Registry No. 10015; Record No. 56f90bc29cb636d83dbc9a37ee2d878cfce14aaf4ced53e1d1bcfa2da45a03e7; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10015/osworld.svg" alt="GPT-6.1 Sol: OSWorld" height="24"> | 10015 |
| OpenAI | GPT-6.1 Sol (max) | 71.4% | openai.com for GPT-6.1 Sol (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 71.4% (max). Registry No. 10015; Record No. 9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26#BR-9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26,
title = {GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 71.4\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26\#BR-9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26},
urldate = {YYYY-MM-DD},
note = {Registry No. 10015; Record No. 9d72832b5e6891857172def6e008f33b1a507ac49b71ac07d234ca1fec117d26; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10015/osworld.svg" alt="GPT-6.1 Sol: OSWorld" height="24"> | 10015 |
| OpenAI | GPT-6.1 Sol (low) | 59.0% | openai.com for GPT-6.1 Sol (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 59.0% (low). Registry No. 10015; Record No. a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533#BR-a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533,
title = {GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 59.0\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533\#BR-a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533},
urldate = {YYYY-MM-DD},
note = {Registry No. 10015; Record No. a8169dd01d977745bd836103e13d82db44ad506645e770996e952350525b8533; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10015/osworld.svg" alt="GPT-6.1 Sol: OSWorld" height="24"> | 10015 |
| OpenAI | GPT-6.1 Sol (medium) | 66.8% | openai.com for GPT-6.1 Sol (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 66.8% (medium). Registry No. 10015; Record No. f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3#BR-f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3,
title = {GPT-6.1 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 66.8\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3\#BR-f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3},
urldate = {YYYY-MM-DD},
note = {Registry No. 10015; Record No. f977e57ee33a052c04145360f49a86a748d944a0b0f7d23f16b86bed291113d3; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10015/osworld.svg" alt="GPT-6.1 Sol: OSWorld" height="24"> | 10015 |
| OpenAI | GPT-6 Astra (low) | 62.2% | openai.com for GPT-6 Astra (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 62.2% (low). Registry No. 10005; Record No. 19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370#BR-19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 62.2\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370\#BR-19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. 19edc871fdb966ed6cb8b609bb8cd9f94eb1eee797e3a523cfb8020714c99370; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |
| OpenAI | GPT-6 Astra (max) | 73.5% | openai.com for GPT-6 Astra (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 73.5% (max). Registry No. 10005; Record No. 3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded#BR-3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 73.5\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded\#BR-3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. 3d38a19a99fec7bf4f9edd6f0a0094505f0726531caea651e689bf0b289e3ded; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |
| OpenAI | GPT-6 Astra (medium) | 69.3% | openai.com for GPT-6 Astra (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.3% (medium). Registry No. 10005; Record No. 80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da#BR-80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 69.3\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da\#BR-80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. 80e4d73ad29047376f97e617a4d18b54fa51814ab690f9066623d7f7128b71da; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |
| OpenAI | GPT-6 Astra (high) | 70.0% | openai.com for GPT-6 Astra (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 70.0% (high). Registry No. 10005; Record No. c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e#BR-c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 70.0\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e\#BR-c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. c605ae83f7632c686c70ef0055b8e0ac60cc76a27cf5cdddfdef294dca4a0c3e; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |
| OpenAI | GPT-6 Astra (xhigh) | 71.3% | openai.com for GPT-6 Astra (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 71.3% (xhigh). Registry No. 10005; Record No. cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3#BR-cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 71.3\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3\#BR-cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. cc1a44dcf3c184c533970838b9df24a2370ae3bc26ef35375d714b7c0ce769d3; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |
| OpenAI | GPT-5.6 Luna (max) | 52.7% | openai.com for GPT-5.6 Luna (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 52.7% (max). Registry No. 10009; Record No. 0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444#BR-0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444,
title = {GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 52.7\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444\#BR-0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444},
urldate = {YYYY-MM-DD},
note = {Registry No. 10009; Record No. 0168392e77f695041245efbc05061b2a5292a0e859ac7ba0c34388efcd00f444; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10009/osworld.svg" alt="GPT-5.6 Luna: OSWorld" height="24"> | 10009 |
| OpenAI | GPT-5.6 Luna (high) | 35.9% | openai.com for GPT-5.6 Luna (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 35.9% (high). Registry No. 10009; Record No. 16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719#BR-16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719,
title = {GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 35.9\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719\#BR-16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719},
urldate = {YYYY-MM-DD},
note = {Registry No. 10009; Record No. 16c9eeff018618514287cac58cd73564cfb8a51a6265a519e471ad5647122719; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10009/osworld.svg" alt="GPT-5.6 Luna: OSWorld" height="24"> | 10009 |
| OpenAI | GPT-5.6 Luna (medium) | 22.7% | openai.com for GPT-5.6 Luna (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 22.7% (medium). Registry No. 10009; Record No. 5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8#BR-5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8,
title = {GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 22.7\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8\#BR-5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8},
urldate = {YYYY-MM-DD},
note = {Registry No. 10009; Record No. 5142833a039e05c580831dd02a9624c8d5de76732d3302e706c66b1e6763a5e8; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10009/osworld.svg" alt="GPT-5.6 Luna: OSWorld" height="24"> | 10009 |
| OpenAI | GPT-5.6 Luna (xhigh) | 48.0% | openai.com for GPT-5.6 Luna (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 48.0% (xhigh). Registry No. 10009; Record No. b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a#BR-b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a,
title = {GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 48.0\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a\#BR-b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a},
urldate = {YYYY-MM-DD},
note = {Registry No. 10009; Record No. b5675ee7a6b3717ee2567c62241fdfa958105d3f6bf68d61ede6c8432eeea20a; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10009/osworld.svg" alt="GPT-5.6 Luna: OSWorld" height="24"> | 10009 |
| OpenAI | GPT-5.6 Luna (low) | 12.0% | openai.com for GPT-5.6 Luna (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 12.0% (low). Registry No. 10009; Record No. daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561#BR-daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561,
title = {GPT-5.6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 12.0\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561\#BR-daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561},
urldate = {YYYY-MM-DD},
note = {Registry No. 10009; Record No. daf075785674cd7c406e839e8987c0fc9404b9c0986d087a05cca7ad40fa9561; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10009/osworld.svg" alt="GPT-5.6 Luna: OSWorld" height="24"> | 10009 |
| OpenAI | GPT-5.6 Sol (high) | 56.5% | openai.com for GPT-5.6 Sol (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 56.5% (high). Registry No. 10010; Record No. 11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf#BR-11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf,
title = {GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 56.5\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf\#BR-11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf},
urldate = {YYYY-MM-DD},
note = {Registry No. 10010; Record No. 11e9a8d71386a02ad3db37912bbbade20ebe73cfb832b89bfeeaa2fd3d773bcf; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10010/osworld.svg" alt="GPT-5.6 Sol: OSWorld" height="24"> | 10010 |
| OpenAI | GPT-5.6 Sol (max) | 66.2% | openai.com for GPT-5.6 Sol (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 66.2% (max). Registry No. 10010; Record No. 2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444#BR-2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444,
title = {GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 66.2\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444\#BR-2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444},
urldate = {YYYY-MM-DD},
note = {Registry No. 10010; Record No. 2a1097f5e682cc8db8e9b038f356ee22c9b1825c15aa1a6aad879c2a16a95444; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10010/osworld.svg" alt="GPT-5.6 Sol: OSWorld" height="24"> | 10010 |
| OpenAI | GPT-5.6 Sol (low) | 29.8% | openai.com for GPT-5.6 Sol (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 29.8% (low). Registry No. 10010; Record No. 4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80#BR-4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80,
title = {GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 29.8\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80\#BR-4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80},
urldate = {YYYY-MM-DD},
note = {Registry No. 10010; Record No. 4e17f0725d38a7b7181221fdaa266c481d0390c717a300b07df2d0a723801a80; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10010/osworld.svg" alt="GPT-5.6 Sol: OSWorld" height="24"> | 10010 |
| OpenAI | GPT-5.6 Sol (medium) | 49.7% | openai.com for GPT-5.6 Sol (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 49.7% (medium). Registry No. 10010; Record No. e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315#BR-e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315,
title = {GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 49.7\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315\#BR-e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315},
urldate = {YYYY-MM-DD},
note = {Registry No. 10010; Record No. e05126b5b0ce24a8e6f9febd801202c67ccefd8ff00493523a412e00504f1315; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10010/osworld.svg" alt="GPT-5.6 Sol: OSWorld" height="24"> | 10010 |
| OpenAI | GPT-5.6 Sol (xhigh) | 60.9% | openai.com for GPT-5.6 Sol (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 60.9% (xhigh). Registry No. 10010; Record No. ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22#BR-ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22,
title = {GPT-5.6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 60.9\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22\#BR-ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22},
urldate = {YYYY-MM-DD},
note = {Registry No. 10010; Record No. ec7e4cb28a2d50fb611148e4a08f836646187b125d1c86b4e82c1ab7a2f4aa22; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10010/osworld.svg" alt="GPT-5.6 Sol: OSWorld" height="24"> | 10010 |
| OpenAI | GPT-6 Luna (xhigh) | 46.7% | openai.com for GPT-6 Luna (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 46.7% (xhigh). Registry No. 10013; Record No. 1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c#BR-1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c,
title = {GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 46.7\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c\#BR-1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c},
urldate = {YYYY-MM-DD},
note = {Registry No. 10013; Record No. 1af2166f4bc4d9daa3144310e87965cca1ac33e46b5021398cc4cbb720d74a6c; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10013/osworld.svg" alt="GPT-6 Luna: OSWorld" height="24"> | 10013 |
| OpenAI | GPT-6 Luna (medium) | 31.5% | openai.com for GPT-6 Luna (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 31.5% (medium). Registry No. 10013; Record No. 4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd#BR-4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd,
title = {GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 31.5\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd\#BR-4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd},
urldate = {YYYY-MM-DD},
note = {Registry No. 10013; Record No. 4f95d0849d3619b4814a5ba29a587bb4176ef973edf642f9ee9703b89cb2f4bd; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10013/osworld.svg" alt="GPT-6 Luna: OSWorld" height="24"> | 10013 |
| OpenAI | GPT-6 Luna (max) | 52.7% | openai.com for GPT-6 Luna (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 52.7% (max). Registry No. 10013; Record No. 57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49#BR-57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49,
title = {GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 52.7\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49\#BR-57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49},
urldate = {YYYY-MM-DD},
note = {Registry No. 10013; Record No. 57f5bdb033ab19aa188816ece50706c593d56784cbf502de083bcca8acddfa49; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10013/osworld.svg" alt="GPT-6 Luna: OSWorld" height="24"> | 10013 |
| OpenAI | GPT-6 Luna (high) | 41.4% | openai.com for GPT-6 Luna (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 41.4% (high). Registry No. 10013; Record No. f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430#BR-f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430,
title = {GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 41.4\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430\#BR-f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430},
urldate = {YYYY-MM-DD},
note = {Registry No. 10013; Record No. f22509c547cffbea5aaae7ba3b90ce14cf96d9af8b6116afcae24026e3c28430; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10013/osworld.svg" alt="GPT-6 Luna: OSWorld" height="24"> | 10013 |
| OpenAI | GPT-6 Luna (low) | 8.3% | openai.com for GPT-6 Luna (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 8.3% (low). Registry No. 10013; Record No. f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e#BR-f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e,
title = {GPT-6 Luna: OSWorld 2.0 v2026.08.08 Offline — Partial, 8.3\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e\#BR-f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e},
urldate = {YYYY-MM-DD},
note = {Registry No. 10013; Record No. f5c0eb4e3fc8091fb5c21bb2d4f1f9c14bbac0b1ecec622d0f657ef0b732b61e; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10013/osworld.svg" alt="GPT-6 Luna: OSWorld" height="24"> | 10013 |
| OpenAI | GPT-6 Sol (xhigh) | 60.5% | openai.com for GPT-6 Sol (xhigh) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 60.5% (xhigh). Registry No. 10014; Record No. 493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f#BR-493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f,
title = {GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 60.5\% (xhigh)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f\#BR-493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f},
urldate = {YYYY-MM-DD},
note = {Registry No. 10014; Record No. 493083de6115a7e72b8bfaaf45a9c346cf68f4e2ac8a0d33d4ecad353dec2b0f; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10014/osworld.svg" alt="GPT-6 Sol: OSWorld" height="24"> | 10014 |
| OpenAI | GPT-6 Sol (max) | 64.4% | openai.com for GPT-6 Sol (max) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 64.4% (max). Registry No. 10014; Record No. 5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d#BR-5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d,
title = {GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 64.4\% (max)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d\#BR-5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d},
urldate = {YYYY-MM-DD},
note = {Registry No. 10014; Record No. 5a923131f54cd5ef55039d5ab5f475650dd2d3e9b4c8155c6ecfe1cb225e054d; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10014/osworld.svg" alt="GPT-6 Sol: OSWorld" height="24"> | 10014 |
| OpenAI | GPT-6 Sol (medium) | 54.0% | openai.com for GPT-6 Sol (medium) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 54.0% (medium). Registry No. 10014; Record No. 8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad#BR-8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad,
title = {GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 54.0\% (medium)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad\#BR-8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad},
urldate = {YYYY-MM-DD},
note = {Registry No. 10014; Record No. 8235bc7380cce0ab80c1339fbcbe34bcdca5293a43a5524ac2677d4108072cad; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10014/osworld.svg" alt="GPT-6 Sol: OSWorld" height="24"> | 10014 |
| OpenAI | GPT-6 Sol (high) | 58.3% | openai.com for GPT-6 Sol (high) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 58.3% (high). Registry No. 10014; Record No. a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838#BR-a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838,
title = {GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 58.3\% (high)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838\#BR-a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838},
urldate = {YYYY-MM-DD},
note = {Registry No. 10014; Record No. a0cea2d08be3a983cb9ecb311fe68f7bced8e6f29c20570f87e7d022cbae4838; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10014/osworld.svg" alt="GPT-6 Sol: OSWorld" height="24"> | 10014 |
| OpenAI | GPT-6 Sol (low) | 43.9% | openai.com for GPT-6 Sol (low) on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 43.9% (low). Registry No. 10014; Record No. e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce#BR-e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce,
title = {GPT-6 Sol: OSWorld 2.0 v2026.08.08 Offline — Partial, 43.9\% (low)},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce\#BR-e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce},
urldate = {YYYY-MM-DD},
note = {Registry No. 10014; Record No. e1144234ab41c394cc1e0dd69ac7ef789ff867f4ca76ef5b439400021c8aa5ce; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10014/osworld.svg" alt="GPT-6 Sol: OSWorld" height="24"> | 10014 |
| OpenAI | GPT-6 Astra | 72.6% | openai.com for GPT-6 Astra on OSWorld 2.0 v2026.08.08 Offline — Partial (opens in a new tab)OSWorld Partial ScoreReported OpenAIReport an issueCiteFill in the access date before citing. Plain text Benchmark Registry. GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 72.6%. Registry No. 10005; Record No. e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205. https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history&result=e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205#BR-e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205. Accessed [YYYY-MM-DD]. Data license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: Benchmark Registry, a project of Densa Labs — https://benchmarkregistry.org/ (CC BY 4.0). Include the cited record URL and identify any changes. BibTeX @misc{br_e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205,
title = {GPT-6 Astra: OSWorld 2.0 v2026.08.08 Offline — Partial, 72.6\%},
publisher = {Benchmark Registry},
url = {https://benchmarkregistry.org/benchmarks/osworld/2.0-v2026.08.08-offline-partial?view=history\&result=e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205\#BR-e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205},
urldate = {YYYY-MM-DD},
note = {Registry No. 10005; Record No. e6a65a690a3e46c6cef1c4404ec73b6c301516b63062c5c3a52f2b6d43b07205; data license: CC BY 4.0; access date to be filled by the reader}
}Badge for this model's latest reported score in the benchmark family; it may differ from this record. Markdown  HTML <img src="https://benchmarkregistry.org/badge/10005/osworld.svg" alt="GPT-6 Astra: OSWorld" height="24"> | 10005 |