# Desktop UI Navigation on OSWorld 50 easy tasks 1.0 (test) Source: [Desktop UI Navigation on OSWorld 50 easy tasks 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:23:50.470Z Evaluation of agent performance on Desktop UI Navigation tasks using the test split of the OSWorld 50 easy tasks 1.0 dataset, measuring success rates such as ASR, SR, and OSR. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Desktop UI Navigation | | Dataset (including any stated split) | OSWorld 50 easy tasks 1.0 (test) | | Metric | ASR | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [ASR](): Higher is better; unit: PERCENT. - [SR](): Higher is better; unit: PERCENT. - [OSR](): Higher is better; unit: PERCENT. ## Recorded results | Method | ASR | Result date | Report paper | | --- | --- | --- | --- | | Claude 3.5 Sonnet | 100 | Not recorded | 68fa78c41e209131cea7a616 | | Claude 3.5 Sonnet v2 | 96 | Not recorded | 68fa78c41e209131cea7a616 | | GPT-4o | 95.8 | Not recorded | 68fa78c41e209131cea7a616 | | Claude 3.5 Sonnet | 95.3 | Not recorded | 68fa78c41e209131cea7a616 | | Claude 3.5 Sonnet v2 | 94.8 | Not recorded | 68fa78c41e209131cea7a616 | | GPT-4-Turbo | 93.3 | Not recorded | 68fa78c41e209131cea7a616 | | GPT-4-Turbo | 91.8 | Not recorded | 68fa78c41e209131cea7a616 | | GPT-4o | 91.2 | Not recorded | 68fa78c41e209131cea7a616 | | Gemini 1.5 | 88.7 | Not recorded | 68fa78c41e209131cea7a616 | | Gemini 1.5 | 80 | Not recorded | 68fa78c41e209131cea7a616 | ### Claude 3.5 Sonnet ASR: 100. Result ID: 6995a1583badfaffe69e0f3c. [Reporting paper]() Recorded evaluation configuration: {"Observation Modality":"Screenshot","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### Claude 3.5 Sonnet v2 ASR: 96. Result ID: 6995a1583badfaffe69e0f3d. [Reporting paper]() Recorded evaluation configuration: {"Observation Modality":"Screenshot","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### GPT-4o ASR: 95.8. Result ID: 6995a1583badfaffe69e0f3a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Observation Modality":"Screenshot","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### Claude 3.5 Sonnet ASR: 95.3. Result ID: 6995a1583badfaffe69e0f41. [Reporting paper]() Recorded evaluation configuration: {"Observation Modality":"SoM","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### Claude 3.5 Sonnet v2 ASR: 94.8. Result ID: 6995a1583badfaffe69e0f42. [Reporting paper]() Recorded evaluation configuration: {"Observation Modality":"SoM","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### GPT-4-Turbo ASR: 93.3. Result ID: 6995a1583badfaffe69e0f39. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Observation Modality":"Screenshot","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### GPT-4-Turbo ASR: 91.8. Result ID: 6995a1583badfaffe69e0f3e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Observation Modality":"SoM","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### GPT-4o ASR: 91.2. Result ID: 6995a1583badfaffe69e0f3f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Observation Modality":"SoM","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### Gemini 1.5 ASR: 88.7. Result ID: 6995a1583badfaffe69e0f40. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Observation Modality":"SoM","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. ### Gemini 1.5 ASR: 80. Result ID: 6995a1583badfaffe69e0f3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Observation Modality":"Screenshot","Action Step Limit":15,"Decoding Temperature":0}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.