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.

73.4% 47/64 of 64 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.

66.2 / 100
Cost $3.36
Tokens 112k / 112k input / output, all attempts
Duration 3h 4m
Filter:
Per-task results for run f4f5d5f3-0156-4b13-8bb9-e86a7b965bec
TaskDifficultyAttemptScoreTestsCompileDurationTokens
CG-AL-E001easy1100.0 / 1007/7OK2m 36s418 / 433
CG-AL-E002easy1100.0 / 1006/6OK2m 30s433 / 378
CG-AL-E003easy1100.0 / 1005/5OK2m 36s410 / 166
CG-AL-E004easy1100.0 / 1006/6OK2m 37s470 / 367
CG-AL-E005easy1100.0 / 10013/13OK2m 43s491 / 651
CG-AL-E006easy1100.0 / 1007/7OK2m 49s470 / 593
CG-AL-E007easy1100.0 / 1007/7OK2m 52s558 / 242
CG-AL-E008easy1100.0 / 1006/6OK2m 53s470 / 468
CG-AL-E009easy1100.0 / 1005/5OK2m 56s442 / 230
CG-AL-E010easy1100.0 / 1005/5OK2m 56s467 / 188
CG-AL-E031easy1100.0 / 1003/3OK3m 4s678 / 659
CG-AL-E032easy1100.0 / 1001/1OK3m 0s377 / 89
CG-AL-E045easy1100.0 / 1004/4OK2m 32s565 / 574
CG-AL-E050easy2100.0 / 10013/13OK2m 29s3.0k / 611
CG-AL-E051easy1100.0 / 10015/15OK2m 37s776 / 1.2k
CG-AL-E052easy2100.0 / 10016/16OK2m 29s1.6k / 763
CG-AL-E053easy1100.0 / 1003/3OK2m 33s461 / 176
CG-AL-E054easy2100.0 / 1009/9OK2m 27s1.7k / 708
CG-AL-E055easy1100.0 / 1008/8OK2m 28s566 / 306
CG-AL-H001hard1100.0 / 10025/25OK2m 36s726 / 436
CG-AL-H002hard1100.0 / 1004/4OK2m 53s747 / 756
CG-AL-H003hard262.5 / 1004/5OK2m 29s2.4k / 1.8k
CG-AL-H004hard1100.0 / 10014/14OK2m 49s779 / 611
CG-AL-H005hard262.5 / 1002/5OK2m 33s3.0k / 2.6k
CG-AL-H006hard1100.0 / 1006/6OK2m 41s715 / 365
CG-AL-H007hard1100.0 / 10010/10OK2m 33s952 / 939
CG-AL-H008hard1100.0 / 10010/10OK2m 29s886 / 586
CG-AL-H009hard1100.0 / 10011/11OK2m 33s1.0k / 1.3k
CG-AL-H010hard1100.0 / 1008/8OK2m 26s901 / 632
CG-AL-H011hard20.0 / 1000/0FAIL14.7s1.5k / 559
CG-AL-H013hard1100.0 / 1009/9OK2m 30s607 / 453
CG-AL-H014hard20.0 / 1000/0FAIL16.1s2.4k / 1.4k
CG-AL-H015hard1100.0 / 1004/4OK2m 27s663 / 449
CG-AL-H016hard20.0 / 1000/0FAIL15.1s2.1k / 1.1k
CG-AL-H017hard20.0 / 1000/0FAIL12.8s1.6k / 708
CG-AL-H018hard1100.0 / 1006/6OK2m 28s732 / 809
CG-AL-H019hard1100.0 / 1005/5OK2m 24s533 / 238
CG-AL-H020hard2100.0 / 10010/10OK2m 27s2.8k / 2.1k
CG-AL-H021hard262.5 / 1006/20OK2m 34s3.3k / 2.2k
CG-AL-H022hard1100.0 / 10021/21OK2m 42s1.2k / 1.6k
CG-AL-H023hard20.0 / 1000/0FAIL2m 23s21k / 22k
CG-AL-H024hard1100.0 / 1009/9OK2m 32s852 / 1.3k
CG-AL-H025hard1100.0 / 1007/7OK2m 28s575 / 302
CG-AL-H026hard20.0 / 1000/0FAIL19.6s3.3k / 1.1k
CG-AL-H205hard1100.0 / 1006/6OK2m 33s784 / 664
CG-AL-M001medium262.5 / 1004/10OK2m 38s2.8k / 3.0k
CG-AL-M002medium1100.0 / 10022/22OK2m 34s730 / 850
CG-AL-M003medium262.5 / 1008/9OK2m 42s2.5k / 2.2k
CG-AL-M004medium2100.0 / 10012/12OK2m 45s2.8k / 3.0k
CG-AL-M005medium262.5 / 10012/22OK2m 53s3.5k / 6.5k
CG-AL-M006medium1100.0 / 10018/18OK2m 47s710 / 1.9k
CG-AL-M007medium2100.0 / 10015/15OK2m 57s3.8k / 7.7k
CG-AL-M008medium262.5 / 10011/12OK2m 51s3.6k / 5.7k
CG-AL-M009medium1100.0 / 10011/11OK2m 50s751 / 2.5k
CG-AL-M010medium262.5 / 10019/21OK2m 58s5.0k / 8.1k
CG-AL-M020medium2100.0 / 10017/17OK2m 34s3.0k / 2.0k
CG-AL-M021medium20.0 / 1000/0FAIL31.2s3.9k / 4.7k
CG-AL-M022medium1100.0 / 1009/9OK2m 40s815 / 1.1k
CG-AL-M023medium1100.0 / 10011/11OK2m 37s808 / 730
CG-AL-M024medium1100.0 / 10010/10OK2m 42s762 / 1.4k
CG-AL-M025medium1100.0 / 1007/7OK2m 34s660 / 502
CG-AL-M026medium20.0 / 1000/0FAIL14.1s2.3k / 1.3k
CG-AL-M088medium1100.0 / 1006/6OK2m 36s678 / 643
CG-AL-M112medium262.5 / 1002/4OK2m 33s2.9k / 2.2k