# OS agent GUI interaction on VeriOS-Bench Source: [OS agent GUI interaction on VeriOS-Bench]() Collection index: [llms.txt]() Data updated: 2026-04-06T07:15:02.375Z Evaluation of agent performance on operating system GUI interaction tasks using the VeriOS-Bench dataset, measured by MC, IM, EA, SA, NS, and Total scores. 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 | OS agent GUI interaction | | Dataset (including any stated split) | VeriOS-Bench | | Metric | MC Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [MC Score](): Higher is better; unit: PERCENT. - [IM Score](): Higher is better; unit: PERCENT. - [EA Score](): Higher is better; unit: PERCENT. - [SA Score](): Higher is better; unit: PERCENT. - [NS Score](): Higher is better; unit: PERCENT. - [Total Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | MC Score | Result date | Report paper | | --- | --- | --- | --- | | VeriOS-Agent | 47.62 | Not recorded | 68fa7b691e209131cea8842b | | VeriOS-Agent | 46.43 | Not recorded | 68fa7b691e209131cea8842b | | VeriOS-Agent | 45.24 | Not recorded | 68fa7b691e209131cea8842b | | VeriOS-Agent | 45.24 | Not recorded | 68fa7b691e209131cea8842b | | UI-TARS-1.5 | 23.81 | Not recorded | 68fa7b691e209131cea8842b | | GLM-4.5V | 23.81 | Not recorded | 68fa7b691e209131cea8842b | | UI-TARS | 21.43 | Not recorded | 68fa7b691e209131cea8842b | | OS-Atlas-Pro | 19.05 | Not recorded | 68fa7b691e209131cea8842b | | UI-TARS | 19.05 | Not recorded | 68fa7b691e209131cea8842b | | UI-TARS | 16.67 | Not recorded | 68fa7b691e209131cea8842b | | Qwen2.5-VL | 16.67 | Not recorded | 68fa7b691e209131cea8842b | | GUI-Owl | 14.29 | Not recorded | 68fa7b691e209131cea8842b | | GPT-4o | 14.29 | Not recorded | 68fa7b691e209131cea8842b | | Qwen-VL-max | 11.9 | Not recorded | 68fa7b691e209131cea8842b | | Qwen2.5-VL | 9.52 | Not recorded | 68fa7b691e209131cea8842b | | Qwen2.5-VL | 2.38 | Not recorded | 68fa7b691e209131cea8842b | | Qwen2.5-VL | 0 | Not recorded | 68fa7b691e209131cea8842b | ### VeriOS-Agent MC Score: 47.62. Result ID: 69d35d6947f34ce924bab955. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"32B","Training Protocol":"RL"}. This record does not assert independent reproduction. ### VeriOS-Agent MC Score: 46.43. Result ID: 69d35d6947f34ce924bab954. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Training Protocol":"RL"}. This record does not assert independent reproduction. ### VeriOS-Agent MC Score: 45.24. Result ID: 69d35d6947f34ce924bab952. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### VeriOS-Agent MC Score: 45.24. Result ID: 69d35d6947f34ce924bab953. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"32B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### UI-TARS-1.5 MC Score: 23.81. Result ID: 69d35d6947f34ce924bab94b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### GLM-4.5V MC Score: 23.81. Result ID: 69d35d6947f34ce924bab951. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### UI-TARS MC Score: 21.43. Result ID: 69d35d6947f34ce924bab949. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### OS-Atlas-Pro MC Score: 19.05. Result ID: 69d35d6947f34ce924bab948. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### UI-TARS MC Score: 19.05. Result ID: 69d35d6947f34ce924bab94a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B","Training Protocol":"DPO"}. This record does not assert independent reproduction. ### UI-TARS MC Score: 16.67. Result ID: 69d35d6947f34ce924bab945. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"2B","Training Protocol":"SFT"}. This record does not assert independent reproduction. ### Qwen2.5-VL MC Score: 16.67. Result ID: 69d35d6947f34ce924bab94e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"72B"}. This record does not assert independent reproduction. ### GUI-Owl MC Score: 14.29. Result ID: 69d35d6947f34ce924bab94d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"32B"}. This record does not assert independent reproduction. ### GPT-4o MC Score: 14.29. Result ID: 69d35d6947f34ce924bab94f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen-VL-max MC Score: 11.9. Result ID: 69d35d6947f34ce924bab950. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL MC Score: 9.52. Result ID: 69d35d6947f34ce924bab94c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"32B"}. This record does not assert independent reproduction. ### Qwen2.5-VL MC Score: 2.38. Result ID: 69d35d6947f34ce924bab947. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL MC Score: 0. Result ID: 69d35d6947f34ce924bab946. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Size":"3B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.