Find a recreational adult sports league volleyball kickballSports
How each model performed on this task. Every chart states its N, and the reward carries a 95% confidence interval once there is more than one run to measure.
Core metrics
Check Pass %
Mean share of the task's checks passing, averaged over runs. N = 1 run per model.
100%
Claude Opus 5.5
100%
GPT-6 sol
83%
Claude Opus 5
33%
GPT-5.6 sol
Extended metrics
Median Time
Wall-clock minutes, median across runs. N = 1 run per model.
16m
Claude Opus 5
4m
GPT-5.6 sol
Tool Calls
Tool invocations, median across runs — counted from the trajectories. N = 1 run per model.
81
Claude Opus 5
28
Claude Opus 5.5
28
GPT-6 sol
19
GPT-5.6 sol
Verifier × model pass rate
How often each verifier passed, per model. Green is passing, red is failing — a red band across a row is a verifier nothing clears.
| Verifier | Claude Opus 5.5n=1 | GPT-6 soln=1 | GPT-5.6 soln=1 | Claude Opus 5n=1 |
|---|---|---|---|---|
| checks (6) | ||||
| 100% | 100% | 0% | 100% | |
How to passConfirm preferred sport(s), skill level (beginner/social), day/time availability, and budget per season ($100). | ||||
| 100% | 100% | 100% | 100% | |
How to passGo to UISP Genova (or the local city's parks & recreation adult sports page if UISP isn't available), and record available leagues matching the criteria, including season length and cost. If ZogSports-style national platforms don't operate in the destination city, use the closest local equivalent (e.g. a national/regional sports-promotion body). | ||||
| 100% | 100% | 0% | 0% | |
How to passSearch for a second independent source confirming the same league details (e.g. a CONI-affiliated club, a regional sports federation page, or a different UISP-affiliated venue), to cross-check the details found in Step 2. | ||||
| 100% | 100% | 0% | 100% | |
How to passSearch for one more local option (a community sports club or a regional league site), recording the same details. | ||||
| 100% | 100% | 0% | 100% | |
How to passFor each shortlisted league (as many as qualify, up to three), confirm it explicitly offers a beginner/social tier, not just competitive divisions. | ||||
| 100% | 100% | 100% | 100% | |
How to passNote whether registration is as an individual (team assigned) or requires bringing a full team, since this affects who can join solo. | ||||
Failure modes on this task
5 tagged verifier failures, by the stage they were assigned to. Higher is worse.
| Failure stage | Claude Opus 5.5 | GPT-6 sol | GPT-5.6 sol | Claude Opus 5 |
|---|---|---|---|---|
| Standing up the evidence1 mode | — | — | 2 | 1 |
| No signal in the judge’s words1 mode · the grader | — | — | 2 | — |