# GUI Agent Planning and Execution on AMEX (test) Source: [GUI Agent Planning and Execution on AMEX (test)]() Collection index: [llms.txt]() Data updated: 2026-03-24T12:34:38.267Z Evaluation of agent performance on GUI Agent Planning and Execution tasks using the AMEX (test) split, reporting success rates across specific applications including Gmail, Booking, Music, SHEIN, News, CM, ToDo, Signal, and Yelp. 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 | GUI Agent Planning and Execution | | Dataset (including any stated split) | AMEX (test) | | Metric | Success Rate (Gmail) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Success Rate (Gmail)](): Higher is better; unit: PERCENT. - [Success Rate (Booking)](): Higher is better; unit: PERCENT. - [Success Rate (Music)](): Higher is better; unit: PERCENT. - [Success Rate (SHEIN)](): Higher is better; unit: PERCENT. - [Success Rate (News)](): Higher is better; unit: PERCENT. - [Success Rate (CM)](): Higher is better; unit: PERCENT. - [Success Rate (ToDo)](): Higher is better; unit: PERCENT. - [Success Rate (Signal)](): Higher is better; unit: PERCENT. - [Success Rate (Yelp)](): Higher is better; unit: PERCENT. - [Overall Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Success Rate (Gmail) | Result date | Report paper | | --- | --- | --- | --- | | MobileIPL | 77.3 | Not recorded | 68fa7ba21e209131cea897a8 | | Qwen2-VL-7B | 75.9 | Not recorded | 68fa7ba21e209131cea897a8 | | UGround-7B | 70.9 | Not recorded | 68fa7ba21e209131cea897a8 | | UI-Tars-7B | 67.7 | Not recorded | 68fa7ba21e209131cea897a8 | | AriaUI-MoE | 63.1 | Not recorded | 68fa7ba21e209131cea897a8 | | SphAgent-7B | 62.4 | Not recorded | 68fa7ba21e209131cea897a8 | | SphAgent-7B | 61.7 | Not recorded | 68fa7ba21e209131cea897a8 | | OS-Atlas-7B | 61.1 | Not recorded | 68fa7ba21e209131cea897a8 | | Qwen2-VL-7B | 58 | Not recorded | 68fa7ba21e209131cea897a8 | | MobileIPL-7B | 57 | Not recorded | 68fa7ba21e209131cea897a8 | | SphAgent-7B | 32.1 | Not recorded | 68fa7ba21e209131cea897a8 | | SeeClick-7B | 28.2 | Not recorded | 68fa7ba21e209131cea897a8 | ### MobileIPL Success Rate (Gmail): 77.3. Result ID: 69c2840147f34ce924aa7713. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"MobileIPL"}. This record does not assert independent reproduction. ### Qwen2-VL-7B Success Rate (Gmail): 75.9. Result ID: 69c2840147f34ce924aa7711. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX + CoaT"}. This record does not assert independent reproduction. ### UGround-7B Success Rate (Gmail): 70.9. Result ID: 69c2840147f34ce924aa770c. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### UI-Tars-7B Success Rate (Gmail): 67.7. Result ID: 69c2840147f34ce924aa770f. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### AriaUI-MoE Success Rate (Gmail): 63.1. Result ID: 69c2840147f34ce924aa770b. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### SphAgent-7B Success Rate (Gmail): 62.4. Result ID: 69c2840147f34ce924aa770d. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AITW + AMEX"}. This record does not assert independent reproduction. ### SphAgent-7B Success Rate (Gmail): 61.7. Result ID: 69c2840147f34ce924aa770a. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### OS-Atlas-7B Success Rate (Gmail): 61.1. Result ID: 69c2840147f34ce924aa770e. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### Qwen2-VL-7B Success Rate (Gmail): 58. Result ID: 69c2840147f34ce924aa7710. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AMEX"}. This record does not assert independent reproduction. ### MobileIPL-7B Success Rate (Gmail): 57. Result ID: 69c2840147f34ce924aa7712. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Data":"AMEX (Seed)"}. This record does not assert independent reproduction. ### SphAgent-7B Success Rate (Gmail): 32.1. Result ID: 69c2840147f34ce924aa7709. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AITW"}. This record does not assert independent reproduction. ### SeeClick-7B Success Rate (Gmail): 28.2. Result ID: 69c2840147f34ce924aa7708. [Reporting paper]() Recorded evaluation configuration: {"Training Data":"AITW+External"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.