Lusor on AndroidWorld (2026-11-09)

SETUP
- Agent: Lusor
- Model: DeepSeek V4.1 Flash (low)
- Seed: 42
- Graders: Mobilerun's (droidrun/android-world c5467e6) and Google's unmodified (google-research/android_world 30583e2)
- Mobilerun times and action counts: Mobilerun published per-task times and action counts
- Mobilerun cost: ESTIMATE: published action count x $0.0766 (mean per-call cost we measured running gpt-6-astra low in our harness); not their bill
- App skills: skills_read = published app skills the agent opened in that task. skill_assisted = the subset with a rule written for an action or reading this task needs (our reading of each skill text); opening a skill whose rules do not apply to the task is not counted as assisted.

RESULTS
- Tasks: 116
- Passed, Mobilerun's grader: 116/116
- Passed, Google's grader: 114/116
- Median time per task: Lusor 48.2 s; Mobilerun 216.9 s (published)
- Median per-task speed ratio: 0.19 (5.2x faster)
- Model cost, all tasks: Lusor $2.95; Mobilerun ~$108.70 (estimate)
- Tool calls: 2,381 (Lusor); published actions: 1,419 (Mobilerun)
- App skills: read in 57 tasks; with rules for the task in 21

TASKS
task | mobilerun_grader | google_grader | lusor_time_s | mobilerun_time_s | speed | lusor_cost_usd
AudioRecorderRecordAudio | pass | pass | 27.2 | 147.0 | 5.4x faster | $0.0082
AudioRecorderRecordAudioWithFileName | pass | fail | 42.3 | 250.9 | 5.9x faster | $0.0132
BrowserDraw | pass | pass | 90.2 | 388.0 | 4.3x faster | $0.0341
BrowserMaze | pass | pass | 51.2 | 350.5 | 6.8x faster | $0.0181
BrowserMultiply | pass | pass | 195.2 | 362.9 | 1.9x faster | $0.0688
CameraTakePhoto | pass | pass | 15.2 | 199.3 | 13.1x faster | $0.0041
CameraTakeVideo | pass | pass | 84.2 | 192.9 | 2.3x faster | $0.0302
ClockStopWatchPausedVerify | pass | pass | 21.2 | 216.7 | 10.2x faster | $0.0086
ClockStopWatchRunning | pass | pass | 18.2 | 122.1 | 6.7x faster | $0.0063
ClockTimerEntry | pass | pass | 21.1 | 117.0 | 5.5x faster | $0.0063
ContactsAddContact | pass | pass | 39.2 | 409.4 | 10.4x faster | $0.0137
ContactsNewContactDraft | pass | pass | 48.2 | 215.0 | 4.5x faster | $0.0191
ExpenseAddMultiple | pass | pass | 72.2 | 471.1 | 6.5x faster | $0.0323
ExpenseAddMultipleFromGallery | pass | pass | 96.2 | 636.8 | 6.6x faster | $0.0394
ExpenseAddMultipleFromMarkor | pass | fail | 126.2 | 318.7 | 2.5x faster | $0.0467
ExpenseAddSingle | pass | pass | 42.2 | 311.7 | 7.4x faster | $0.0169
ExpenseDeleteDuplicates | pass | pass | 42.2 | 418.0 | 9.9x faster | $0.0203
ExpenseDeleteDuplicates2 | pass | pass | 93.3 | 382.5 | 4.1x faster | $0.0303
ExpenseDeleteMultiple | pass | pass | 30.2 | 271.1 | 9.0x faster | $0.0131
ExpenseDeleteMultiple2 | pass | pass | 180.3 | 366.7 | 2.0x faster | $0.0709
ExpenseDeleteSingle | pass | pass | 45.2 | 155.6 | 3.4x faster | $0.0153
FilesDeleteFile | pass | pass | 24.2 | 204.4 | 8.4x faster | $0.0139
FilesMoveFile | pass | pass | 51.2 | 380.8 | 7.4x faster | $0.0220
MarkorAddNoteHeader | pass | pass | 69.2 | 224.2 | 3.2x faster | $0.0309
MarkorChangeNoteContent | pass | pass | 54.2 | 217.5 | 4.0x faster | $0.0238
MarkorCreateFolder | pass | pass | 18.2 | 157.9 | 8.7x faster | $0.0059
MarkorCreateNote | pass | pass | 30.2 | 204.5 | 6.8x faster | $0.0110
MarkorCreateNoteAndSms | pass | pass | 57.2 | 351.5 | 6.1x faster | $0.0175
MarkorCreateNoteFromClipboard | pass | pass | 36.2 | 258.9 | 7.2x faster | $0.0135
MarkorDeleteAllNotes | pass | pass | 48.2 | 198.6 | 4.1x faster | $0.0218
MarkorDeleteNewestNote | pass | pass | 21.2 | 194.0 | 9.2x faster | $0.0088
MarkorDeleteNote | pass | pass | 21.2 | 140.9 | 6.6x faster | $0.0099
MarkorEditNote | pass | pass | 27.3 | 143.1 | 5.2x faster | $0.0087
MarkorMergeNotes | pass | pass | 72.2 | 318.0 | 4.4x faster | $0.0285
MarkorMoveNote | pass | pass | 27.2 | 275.9 | 10.1x faster | $0.0140
MarkorTranscribeReceipt | pass | pass | 45.2 | 233.7 | 5.2x faster | $0.0154
MarkorTranscribeVideo | pass | pass | 354.2 | 489.2 | 1.4x faster | $0.1580
NotesIsTodo | pass | pass | 15.2 | 85.9 | 5.7x faster | $0.0067
NotesMeetingAttendeeCount | pass | pass | 24.2 | 147.0 | 6.1x faster | $0.0109
NotesRecipeIngredientCount | pass | pass | 21.2 | 137.5 | 6.5x faster | $0.0097
NotesTodoItemCount | pass | pass | 30.2 | 156.7 | 5.2x faster | $0.0120
OpenAppTaskEval | pass | pass | 24.2 | 178.4 | 7.4x faster | $0.0090
OsmAndFavorite | pass | pass | 117.2 | 166.8 | 1.4x faster | $0.0482
OsmAndMarker | pass | pass | 36.2 | 212.6 | 5.9x faster | $0.0136
OsmAndTrack | pass | pass | 153.2 | 887.7 | 5.8x faster | $0.0874
RecipeAddMultipleRecipes | pass | pass | 186.2 | 429.7 | 2.3x faster | $0.0537
RecipeAddMultipleRecipesFromImage | pass | pass | 222.3 | 443.9 | 2.0x faster | $0.0602
RecipeAddMultipleRecipesFromMarkor | pass | pass | 198.4 | 482.7 | 2.4x faster | $0.0490
RecipeAddMultipleRecipesFromMarkor2 | pass | pass | 249.2 | 517.5 | 2.1x faster | $0.0677
RecipeAddSingleRecipe | pass | pass | 93.2 | 244.8 | 2.6x faster | $0.0209
RecipeDeleteDuplicateRecipes | pass | pass | 51.2 | 250.3 | 4.9x faster | $0.0155
RecipeDeleteDuplicateRecipes2 | pass | pass | 96.2 | 327.1 | 3.4x faster | $0.0353
RecipeDeleteDuplicateRecipes3 | pass | pass | 240.2 | 620.1 | 2.6x faster | $0.0355
RecipeDeleteMultipleRecipes | pass | pass | 39.2 | 234.1 | 6.0x faster | $0.0138
RecipeDeleteMultipleRecipesWithConstraint | pass | pass | 282.3 | 539.8 | 1.9x faster | $0.0458
RecipeDeleteMultipleRecipesWithNoise | pass | pass | 108.2 | 474.7 | 4.4x faster | $0.0238
RecipeDeleteSingleRecipe | pass | pass | 18.2 | 171.5 | 9.4x faster | $0.0064
RecipeDeleteSingleWithRecipeWithNoise | pass | pass | 48.2 | 301.6 | 6.3x faster | $0.0122
RetroCreatePlaylist | pass | pass | 57.3 | 663.2 | 11.6x faster | $0.0221
RetroPlayingQueue | pass | pass | 51.2 | 768.0 | 15.0x faster | $0.0242
RetroPlaylistDuration | pass | pass | 240.2 | 494.9 | 2.1x faster | $0.0798
RetroSavePlaylist | pass | pass | 84.4 | 603.7 | 7.2x faster | $0.0388
SaveCopyOfReceiptTaskEval | pass | pass | 51.2 | 266.3 | 5.2x faster | $0.0216
SimpleCalendarAddOneEvent | pass | pass | 69.2 | 420.8 | 6.1x faster | $0.0276
SimpleCalendarAddOneEventInTwoWeeks | pass | pass | 69.2 | 394.2 | 5.7x faster | $0.0276
SimpleCalendarAddOneEventRelativeDay | pass | pass | 78.2 | 435.5 | 5.6x faster | $0.0298
SimpleCalendarAddOneEventTomorrow | pass | pass | 72.2 | 328.1 | 4.5x faster | $0.0301
SimpleCalendarAddRepeatingEvent | pass | pass | 78.2 | 478.9 | 6.1x faster | $0.0312
SimpleCalendarAnyEventsOnDate | pass | pass | 21.2 | 106.3 | 5.0x faster | $0.0084
SimpleCalendarDeleteEvents | pass | pass | 42.2 | 261.4 | 6.2x faster | $0.0167
SimpleCalendarDeleteEventsOnRelativeDay | pass | pass | 33.2 | 215.6 | 6.5x faster | $0.0144
SimpleCalendarDeleteOneEvent | pass | pass | 36.2 | 266.2 | 7.4x faster | $0.0128
SimpleCalendarEventOnDateAtTime | pass | pass | 33.2 | 122.3 | 3.7x faster | $0.0122
SimpleCalendarEventsInNextWeek | pass | pass | 93.1 | 150.2 | 1.6x faster | $0.0252
SimpleCalendarEventsInTimeRange | pass | pass | 21.2 | 110.1 | 5.2x faster | $0.0073
SimpleCalendarEventsOnDate | pass | pass | 24.2 | 93.2 | 3.9x faster | $0.0097
SimpleCalendarFirstEventAfterStartTime | pass | pass | 42.2 | 112.8 | 2.7x faster | $0.0142
SimpleCalendarLocationOfEvent | pass | pass | 18.2 | 113.1 | 6.2x faster | $0.0076
SimpleCalendarNextEvent | pass | pass | 27.2 | 116.5 | 4.3x faster | $0.0106
SimpleCalendarNextMeetingWithPerson | pass | pass | 81.2 | 170.3 | 2.1x faster | $0.0379
SimpleDrawProCreateDrawing | pass | pass | 42.2 | 275.3 | 6.5x faster | $0.0171
SimpleSmsReply | pass | pass | 21.2 | 159.6 | 7.5x faster | $0.0053
SimpleSmsReplyMostRecent | pass | pass | 27.2 | 178.5 | 6.6x faster | $0.0072
SimpleSmsResend | pass | pass | 24.2 | 147.4 | 6.1x faster | $0.0083
SimpleSmsSend | pass | pass | 21.2 | 172.9 | 8.2x faster | $0.0062
SimpleSmsSendClipboardContent | pass | pass | 24.7 | 173.3 | 7.0x faster | $0.0085
SimpleSmsSendReceivedAddress | pass | pass | 54.2 | 198.1 | 3.7x faster | $0.0168
SportsTrackerActivitiesCountForWeek | pass | pass | 351.3 | 279.7 | 1.3x slower | $0.0622
SportsTrackerActivitiesOnDate | pass | pass | 72.2 | 399.9 | 5.5x faster | $0.0231
SportsTrackerActivityDuration | pass | pass | 30.2 | 246.9 | 8.2x faster | $0.0115
SportsTrackerLongestDistanceActivity | pass | pass | 153.2 | 239.2 | 1.6x faster | $0.0500
SportsTrackerTotalDistanceForCategoryOverInterval | pass | pass | 144.2 | 364.8 | 2.5x faster | $0.0494
SportsTrackerTotalDurationForCategoryThisWeek | pass | pass | 69.4 | 217.0 | 3.1x faster | $0.0302
SystemBluetoothTurnOff | pass | pass | 27.2 | 125.1 | 4.6x faster | $0.0106
SystemBluetoothTurnOffVerify | pass | pass | 15.2 | 136.9 | 9.0x faster | $0.0057
SystemBluetoothTurnOn | pass | pass | 18.2 | 121.9 | 6.7x faster | $0.0066
SystemBluetoothTurnOnVerify | pass | pass | 24.2 | 108.0 | 4.5x faster | $0.0061
SystemBrightnessMax | pass | pass | 18.2 | 132.0 | 7.3x faster | $0.0068
SystemBrightnessMaxVerify | pass | pass | 27.2 | 125.6 | 4.6x faster | $0.0087
SystemBrightnessMin | pass | pass | 33.2 | 129.7 | 3.9x faster | $0.0121
SystemBrightnessMinVerify | pass | pass | 18.2 | 117.4 | 6.5x faster | $0.0064
SystemCopyToClipboard | pass | pass | 54.8 | 221.3 | 4.0x faster | $0.0209
SystemWifiTurnOff | pass | pass | 24.2 | 155.7 | 6.4x faster | $0.0064
SystemWifiTurnOffVerify | pass | pass | 30.2 | 93.7 | 3.1x faster | $0.0110
SystemWifiTurnOn | pass | pass | 15.2 | 110.6 | 7.3x faster | $0.0054
SystemWifiTurnOnVerify | pass | pass | 21.2 | 97.1 | 4.6x faster | $0.0060
TasksCompletedTasksForDate | pass | pass | 117.3 | 244.0 | 2.1x faster | $0.0428
TasksDueNextWeek | pass | pass | 180.2 | 154.4 | 1.2x slower | $0.0712
TasksDueOnDate | pass | pass | 171.3 | 139.4 | 1.2x slower | $0.0577
TasksHighPriorityTasks | pass | pass | 273.3 | 193.5 | 1.4x slower | $0.0864
TasksHighPriorityTasksDueOnDate | pass | pass | 81.2 | 133.7 | 1.6x faster | $0.0232
TasksIncompleteTasksOnDate | pass | pass | 219.2 | 126.2 | 1.7x slower | $0.0572
TurnOffWifiAndTurnOnBluetooth | pass | pass | 54.2 | 164.8 | 3.0x faster | $0.0191
TurnOnWifiAndOpenApp | pass | pass | 51.2 | 134.1 | 2.6x faster | $0.0175
VlcCreatePlaylist | pass | pass | 75.2 | 331.1 | 4.4x faster | $0.0351
VlcCreateTwoPlaylists | pass | pass | 114.2 | 420.5 | 3.7x faster | $0.0616