Run success rate

Run success rate

Tasks the run solved on its last attempt / tasks attempted in this run.

Formula: COUNT(distinct tasks where last attempt passed) / COUNT(distinct tasks attempted in this run)

Per-run metric for the model's "final answer" on each task. Differs from leaderboard pass_at_n: this denominator is the run's own attempted-task count, not the task set size, so partial runs are not penalised for unattempted tasks.

81.8% 90/110 of 110 tasks attempted in this run
Avg score

Avg attempt score

Mean per-attempt score on a 0–100 point scale (partial credit). Drill-down only.

Formula: Mean of attempt scores across all results rows: SUM(score) / COUNT(*) over the results table. Each attempt earns 0–100 points based on compile + test outcomes.

Drill-down companion to pass_at_n. Rewards partial credit but not directly comparable to pass rate; use for within-model analysis.

71.4 / 100
Cost $4.03
Tokens 186k / 124k input / output, all attempts
Duration 1h 32m
Filter:
Per-task results for run 47caf9d8-1397-4ada-8d2e-7001f7583e0c
TaskDifficultyAttemptScoreTestsCompileDurationTokens
CG-AL-E001easy1100.0 / 1007/7OK38.6s418 / 433
CG-AL-E002easy1100.0 / 1006/6OK3m 36s433 / 378
CG-AL-E003easy1100.0 / 1005/5OK45.9s410 / 166
CG-AL-E004easy1100.0 / 1006/6OK53.6s470 / 367
CG-AL-E005easy1100.0 / 10013/13OK48.4s491 / 534
CG-AL-E006easy1100.0 / 1007/7OK4m 18s470 / 596
CG-AL-E007easy1100.0 / 1007/7OK59.4s558 / 242
CG-AL-E008easy1100.0 / 1006/6OK58.7s470 / 492
CG-AL-E009easy1100.0 / 1005/5OK1m 3s442 / 254
CG-AL-E010easy1100.0 / 1005/5OK1m 8s467 / 188
CG-AL-E031easy1100.0 / 1003/3OK1m 10s678 / 659
CG-AL-E032easy1100.0 / 1001/1OK1m 2s377 / 89
CG-AL-E045easy1100.0 / 1004/4OK24.4s565 / 434
CG-AL-E050easy1100.0 / 10013/13OK23.3s635 / 299
CG-AL-E051easy1100.0 / 10015/15OK30.8s776 / 1.2k
CG-AL-E052easy2100.0 / 10016/16OK23.4s2.4k / 763
CG-AL-E053easy1100.0 / 1003/3OK3m 10s461 / 172
CG-AL-E054easy2100.0 / 1009/9OK21.6s1.7k / 662
CG-AL-E055easy1100.0 / 1008/8OK22.6s566 / 306
CG-AL-E056easy262.5 / 10013/15OK25.0s2.8k / 998
CG-AL-E057easy20.0 / 1000/0FAIL11.4s2.2k / 1.1k
CG-AL-E058easy1100.0 / 1001/1OK20.7s628 / 168
CG-AL-H001hard1100.0 / 10025/25OK25.5s726 / 433
CG-AL-H002hard1100.0 / 1004/4OK28.1s747 / 749
CG-AL-H003hard1100.0 / 1005/5OK28.8s766 / 881
CG-AL-H004hard1100.0 / 10014/14OK26.3s779 / 625
CG-AL-H005hard1100.0 / 1006/6OK27.8s1.2k / 902
CG-AL-H006hard1100.0 / 1006/6OK23.6s715 / 365
CG-AL-H007hard1100.0 / 10010/10OK27.7s952 / 1.1k
CG-AL-H008hard1100.0 / 10010/10OK24.9s886 / 586
CG-AL-H009hard1100.0 / 10011/11OK31.9s1.0k / 1.3k
CG-AL-H010hard1100.0 / 1008/8OK24.5s901 / 632
CG-AL-H011hard20.0 / 1000/0FAIL9.6s1.6k / 560
CG-AL-H013hard1100.0 / 1009/9OK24.1s607 / 464
CG-AL-H014hard20.0 / 1000/0FAIL13.1s2.4k / 1.4k
CG-AL-H015hard1100.0 / 1004/4OK21.2s663 / 449
CG-AL-H016hard1100.0 / 1004/4OK22.7s703 / 402
CG-AL-H017hard20.0 / 1000/0FAIL10.1s1.6k / 742
CG-AL-H018hard1100.0 / 1006/6OK26.8s732 / 808
CG-AL-H019hard1100.0 / 1005/5OK19.6s533 / 238
CG-AL-H020hard2100.0 / 10010/10OK27.3s2.8k / 2.1k
CG-AL-H021hard2100.0 / 10020/20OK25.8s30k / 2.2k
CG-AL-H022hard1100.0 / 10021/21OK29.1s1.2k / 1.4k
CG-AL-H023hard20.0 / 1000/0FAIL2m 27s20k / 22k
CG-AL-H024hard1100.0 / 1009/9OK55.6s852 / 1.3k
CG-AL-H025hard1100.0 / 1007/7OK26.1s575 / 302
CG-AL-H026hard1100.0 / 1008/8OK20.6s750 / 471
CG-AL-H027hard262.5 / 1002/4OK17.4s3.4k / 500
CG-AL-H028hard1100.0 / 1009/9OK23.0s1.1k / 786
CG-AL-H029hard20.0 / 1000/0FAIL22.4s4.6k / 3.7k
CG-AL-H030hard2100.0 / 10010/10OK30.9s4.3k / 3.1k
CG-AL-H031hard1100.0 / 10022/22OK30.7s1.6k / 1.7k
CG-AL-H032hard1100.0 / 10019/19OK29.1s1.3k / 1.4k
CG-AL-H033hard1100.0 / 1005/5OK3m 1s1.2k / 2.1k
CG-AL-H034hard2100.0 / 1003/3OK18.0s2.1k / 465
CG-AL-H035hard1100.0 / 1003/3OK17.8s500 / 200
CG-AL-H036hard2100.0 / 1005/5OK17.7s1.6k / 503
CG-AL-H037hard1100.0 / 1003/3OK18.5s810 / 292
CG-AL-H038hard1100.0 / 1003/3OK18.7s892 / 346
CG-AL-H039hard1100.0 / 1004/4OK21.3s778 / 357
CG-AL-H040hard1100.0 / 1002/2OK18.9s650 / 252
CG-AL-H041hard1100.0 / 1003/3OK24.1s928 / 595
CG-AL-H042hard1100.0 / 1003/3OK17.2s663 / 181
CG-AL-H043hard1100.0 / 1005/5OK18.8s655 / 150
CG-AL-H050hard262.5 / 1002/3OK17.5s2.3k / 555
CG-AL-H051hard1100.0 / 1004/4OK17.3s994 / 196
CG-AL-H052hard1100.0 / 1005/5OK17.3s1.1k / 157
CG-AL-H053hard1100.0 / 1004/4OK18.2s852 / 341
CG-AL-H054hard20.0 / 1000/0FAIL10.4s3.0k / 1.1k
CG-AL-H056hard20.0 / 1000/0FAIL8.1s2.4k / 454
CG-AL-H057hard1100.0 / 1004/4OK2m 36s1.1k / 412
CG-AL-H058hard1100.0 / 1005/5OK19.5s934 / 459
CG-AL-H205hard1100.0 / 1006/6OK20.7s784 / 664
CG-AL-M001medium20.0 / 1000/0FAIL11.7s2.3k / 1.3k
CG-AL-M002medium1100.0 / 10022/22OK22.9s730 / 842
CG-AL-M003medium1100.0 / 1009/9OK28.7s692 / 1.2k
CG-AL-M004medium1100.0 / 10012/12OK2m 54s778 / 1.5k
CG-AL-M005medium1100.0 / 10022/22OK42.7s948 / 2.8k
CG-AL-M006medium1100.0 / 10018/18OK35.2s710 / 1.8k
CG-AL-M007medium262.5 / 10013/15OK1m 58s3.6k / 4.9k
CG-AL-M008medium2100.0 / 10012/12OK40.5s3.9k / 6.4k
CG-AL-M009medium1100.0 / 10011/11OK1m 30s751 / 2.2k
CG-AL-M010medium1100.0 / 10021/21OK3m 28s963 / 2.7k
CG-AL-M020medium2100.0 / 10017/17OK24.0s3.0k / 2.0k
CG-AL-M021medium1100.0 / 10011/11OK23.8s913 / 835
CG-AL-M022medium1100.0 / 1009/9OK25.6s815 / 1.1k
CG-AL-M023medium262.5 / 10010/11OK22.3s2.7k / 1.5k
CG-AL-M024medium1100.0 / 10010/10OK28.2s762 / 1.5k
CG-AL-M025medium1100.0 / 1007/7OK20.6s660 / 524
CG-AL-M026medium262.5 / 1001/8OK20.1s2.1k / 1.4k
CG-AL-M027medium2100.0 / 10018/18OK31.8s6.0k / 4.0k
CG-AL-M028medium2100.0 / 1003/3OK2m 35s1.7k / 902
CG-AL-M029medium2100.0 / 1003/3OK2m 33s1.4k / 318
CG-AL-M031medium20.0 / 1000/0FAIL15.0s3.0k / 2.4k
CG-AL-M032medium10.0 / 1000/0FAIL0ms0 / 0
CG-AL-M033medium1100.0 / 1002/2OK19.5s740 / 390
CG-AL-M034medium262.5 / 1000/2OK17.4s2.6k / 656
CG-AL-M035medium1100.0 / 1002/2OK21.4s549 / 147
CG-AL-M036medium1100.0 / 1002/2OK20.7s915 / 553
CG-AL-M037medium20.0 / 1000/0FAIL8.8s2.1k / 712
CG-AL-M038medium1100.0 / 1002/2OK19.1s643 / 262
CG-AL-M039medium1100.0 / 1002/2OK2m 45s715 / 499
CG-AL-M040medium20.0 / 1000/0FAIL8.0s1.3k / 467
CG-AL-M041medium1100.0 / 1003/3OK18.7s506 / 159
CG-AL-M042medium1100.0 / 1008/8OK19.6s758 / 405
CG-AL-M043medium1100.0 / 1005/5OK20.1s682 / 426
CG-AL-M044medium1100.0 / 1006/6OK2m 39s990 / 812
CG-AL-M045medium1100.0 / 1006/6OK17.6s1.0k / 248
CG-AL-M088medium1100.0 / 1006/6OK20.7s678 / 640
CG-AL-M112medium1100.0 / 1004/4OK26.2s913 / 1.1k