# Mobile GUI Navigation on MobileWorld GUI-Only 117 tasks (test) Source: [Mobile GUI Navigation on MobileWorld GUI-Only 117 tasks (test)]() Collection index: [llms.txt]() Data updated: 2026-06-19T08:07:26.918Z Evaluation of agent performance on Mobile GUI Navigation tasks using the specific 117-task, GUI-Only subset of the MobileWorld dataset (test split). 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 | Mobile GUI Navigation | | Dataset (including any stated split) | MobileWorld GUI-Only 117 tasks (test) | | Metric | Pass@1 Success Rate | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 32 of 32 | | 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 - [Pass@1 Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Pass@1 Success Rate | Result date | Report paper | | --- | --- | --- | --- | | GUI-Owl-1.5-32B-Instruct | 43.9 | Not recorded | 6a34cd40206542251d7c60c2 | | GUI-Owl-1.5-32B | 43.9 | Not recorded | 6a34cd40206542251d7c60c6 | | ForgeOwl-8B | 41 | Not recorded | 6a34cd40206542251d7c60c6 | | MAI-UI-235B-A22B | 39.7 | Not recorded | 6a34cd40206542251d7c60c2 | | MAI-UI-235B-A22B | 39.7 | Not recorded | 6a34cd40206542251d7c60c6 | | GUI-Owl-1.5-8B-Instruct | 38.2 | Not recorded | 6a34cd40206542251d7c60c2 | | GUI-Owl-1.5-8B | 37.6 | Not recorded | 6a34cd40206542251d7c60c6 | | MAI-UI-32B | 36.2 | Not recorded | 6a34cd40206542251d7c60c2 | | MAI-UI-32B | 36.2 | Not recorded | 6a34cd40206542251d7c60c6 | | GUI-Owl-1.5-2B-Instruct | 32.2 | Not recorded | 6a34cd40206542251d7c60c2 | | GUI-Owl-1.5-2B | 32.2 | Not recorded | 6a34cd40206542251d7c60c6 | | MemGUI-Agent-235B | 29.1 | Not recorded | 6a34cd40206542251d7c60c2 | | MAI-UI-8B | 27.5 | Not recorded | 6a34cd40206542251d7c60c2 | | MAI-UI-8B | 27.5 | Not recorded | 6a34cd40206542251d7c60c6 | | MemGUI-8B-SFT | 17.9 | Not recorded | 6a34cd40206542251d7c60c2 | | OpenMobile-8B | 17.7 | Not recorded | 6a34cd40206542251d7c60c2 | | OpenMobile-8B | 17.7 | Not recorded | 6a34cd40206542251d7c60c6 | | ClawGUI-2B | 17.1 | Not recorded | 6a34cd40206542251d7c60c2 | | ClawGUI-2B | 17.1 | Not recorded | 6a34cd40206542251d7c60c6 | | OpenMobile-7B | 14.8 | Not recorded | 6a34cd40206542251d7c60c2 | | OpenMobile-7B | 14.8 | Not recorded | 6a34cd40206542251d7c60c6 | | Qwen3-VL-235B-Thinking | 14.5 | Not recorded | 6a34cd40206542251d7c60c2 | | Qwen3-VL-235B-Thinking | 14.5 | Not recorded | 6a34cd40206542251d7c60c6 | | Qwen3-VL-235B-Instruct | 12.8 | Not recorded | 6a34cd40206542251d7c60c2 | | Qwen3-VL-235B | 12.8 | Not recorded | 6a34cd40206542251d7c60c6 | | Qwen3-VL-32B-Instruct | 11.9 | Not recorded | 6a34cd40206542251d7c60c2 | | Qwen3-VL-32B | 11.9 | Not recorded | 6a34cd40206542251d7c60c6 | | ForgeQwen3-8B | 10.3 | Not recorded | 6a34cd40206542251d7c60c6 | | Qwen3-VL-8B-Instruct | 9.4 | Not recorded | 6a34cd40206542251d7c60c2 | | ScaleCUA-7B | 7.7 | Not recorded | 6a34cd40206542251d7c60c2 | | ScaleCUA-7B | 7.7 | Not recorded | 6a34cd40206542251d7c60c6 | | Qwen3-VL-8B | 7.6 | Not recorded | 6a34cd40206542251d7c60c6 | ### GUI-Owl-1.5-32B-Instruct Pass@1 Success Rate: 43.9. Result ID: 6a34f85ffbc373592c322cd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### GUI-Owl-1.5-32B Pass@1 Success Rate: 43.9. Result ID: 6a34f8bafbc373592c322ea6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### ForgeOwl-8B Pass@1 Success Rate: 41. Result ID: 6a34f8bafbc373592c322eb5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only","Adaptation Data":"AndroidWorld-derived adaptation data only"}. This record does not assert independent reproduction. ### MAI-UI-235B-A22B Pass@1 Success Rate: 39.7. Result ID: 6a34f85ffbc373592c322cd4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### MAI-UI-235B-A22B Pass@1 Success Rate: 39.7. Result ID: 6a34f8bafbc373592c322ea7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### GUI-Owl-1.5-8B-Instruct Pass@1 Success Rate: 38.2. Result ID: 6a34f85ffbc373592c322cd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### GUI-Owl-1.5-8B Pass@1 Success Rate: 37.6. Result ID: 6a34f8bafbc373592c322ea8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### MAI-UI-32B Pass@1 Success Rate: 36.2. Result ID: 6a34f85ffbc373592c322cd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### MAI-UI-32B Pass@1 Success Rate: 36.2. Result ID: 6a34f8bafbc373592c322ea9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### GUI-Owl-1.5-2B-Instruct Pass@1 Success Rate: 32.2. Result ID: 6a34f85ffbc373592c322cd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### GUI-Owl-1.5-2B Pass@1 Success Rate: 32.2. Result ID: 6a34f8bafbc373592c322eaa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### MemGUI-Agent-235B Pass@1 Success Rate: 29.1. Result ID: 6a34f85ffbc373592c322ce1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory Type":"ConAct","Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction. ### MAI-UI-8B Pass@1 Success Rate: 27.5. Result ID: 6a34f85ffbc373592c322cd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### MAI-UI-8B Pass@1 Success Rate: 27.5. Result ID: 6a34f8bafbc373592c322eab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### MemGUI-8B-SFT Pass@1 Success Rate: 17.9. Result ID: 6a34f85ffbc373592c322ce2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory Type":"ConAct","Evaluation Protocol":"SFT"}. This record does not assert independent reproduction. ### OpenMobile-8B Pass@1 Success Rate: 17.7. Result ID: 6a34f85ffbc373592c322cdd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### OpenMobile-8B Pass@1 Success Rate: 17.7. Result ID: 6a34f8bafbc373592c322eb0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### ClawGUI-2B Pass@1 Success Rate: 17.1. Result ID: 6a34f85ffbc373592c322cde. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### ClawGUI-2B Pass@1 Success Rate: 17.1. Result ID: 6a34f8bafbc373592c322eb1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### OpenMobile-7B Pass@1 Success Rate: 14.8. Result ID: 6a34f85ffbc373592c322cdf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### OpenMobile-7B Pass@1 Success Rate: 14.8. Result ID: 6a34f8bafbc373592c322eb2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### Qwen3-VL-235B-Thinking Pass@1 Success Rate: 14.5. Result ID: 6a34f85ffbc373592c322cd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### Qwen3-VL-235B-Thinking Pass@1 Success Rate: 14.5. Result ID: 6a34f8bafbc373592c322eac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### Qwen3-VL-235B-Instruct Pass@1 Success Rate: 12.8. Result ID: 6a34f85ffbc373592c322cda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### Qwen3-VL-235B Pass@1 Success Rate: 12.8. Result ID: 6a34f8bafbc373592c322ead. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### Qwen3-VL-32B-Instruct Pass@1 Success Rate: 11.9. Result ID: 6a34f85ffbc373592c322cdb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### Qwen3-VL-32B Pass@1 Success Rate: 11.9. Result ID: 6a34f8bafbc373592c322eae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### ForgeQwen3-8B Pass@1 Success Rate: 10.3. Result ID: 6a34f8bafbc373592c322eb4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only","Adaptation Data":"AndroidWorld-derived adaptation data only"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-Instruct Pass@1 Success Rate: 9.4. Result ID: 6a34f85ffbc373592c322cdc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### ScaleCUA-7B Pass@1 Success Rate: 7.7. Result ID: 6a34f85ffbc373592c322ce0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Memory Type":"Action-Thought"}. This record does not assert independent reproduction. ### ScaleCUA-7B Pass@1 Success Rate: 7.7. Result ID: 6a34f8bafbc373592c322eb3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Agent Category":"Open-data mobile GUI agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Pass@1 Success Rate: 7.6. Result ID: 6a34f8bafbc373592c322eaf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Agent Category":"Open-weight GUI / VLM agents","Reporting Protocol":"GUI-only"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.