Additional-owner card activation funnelCards & Interchange
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 trial to measure.
Core metrics
Check Pass %
Mean share of the task's checks passing, averaged over trials. N = 10 trials per model.
85%
Claude Opus 4.8
40%
Gemini 3.1 Pro
29%
GPT-5.5
0%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
14m
GPT-5.5
6m
Gemini 3.1 Pro
4m
Claude Opus 4.8
3m
Gemini 3.5 Flash
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
294
GPT-5.5
60
Gemini 3.1 Pro
51
Claude Opus 4.8
24
Gemini 3.5 Flash
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 | GPT-5.5n=10 | Claude Opus 4.8n=10 | Gemini 3.1 Pron=10 | Gemini 3.5 Flashn=10 |
|---|---|---|---|---|
| checks (13) | ||||
| answer_additional_owner_count | 10% | 90% | 30% | 0% |
| answer_never_ordered_count | 20% | 90% | 30% | 0% |
| answer_paid_extra_card_users | 10% | 90% | 30% | 0% |
| answer_virtual_funnel_activation_wallet | 10% | 60% | 30% | 0% |
| answer_virtual_ordered_count | 10% | 90% | 30% | 0% |
| deliverable_jira_ticket_filed | 80% | 100% | 40% | 0% |
| deliverable_slack_channel_exists | 80% | 100% | 100% | 0% |
| deliverable_slack_message_content | 10% | 60% | 30% | 0% |
| floor_all_connectors | 80% | 100% | 100% | 0% |
| floor_sql_investigation | 100% | 100% | 100% | 100% |
| floor_write_back | 80% | 100% | 100% | 0% |
| trace_cohort_derivation_method | 10% | 90% | 30% | 0% |
| trace_trap_avoidance | 30% | 80% | 30% | 90% |