Stuck card auth holds after POS declinesCards & 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.
73%
GPT-5.5
65%
Claude Opus 4.8
55%
Gemini 3.1 Pro
10%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
32m
Gemini 3.5 Flash
4m
Gemini 3.1 Pro
2m
Claude Opus 4.8
2m
GPT-5.5
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
67
Gemini 3.5 Flash
34
Gemini 3.1 Pro
21
GPT-5.5
17
Claude Opus 4.8
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 (12) | ||||
| answer_declined_with_hold_and_stuck_counts | 80% | 50% | 40% | 0% |
| answer_event_window | 60% | 50% | 40% | 0% |
| answer_frozen_and_release_needed | 50% | 50% | 30% | 0% |
| deliverable_jira_ticket_filed | 100% | 90% | 90% | 30% |
| deliverable_slack_channel_exists | 100% | 100% | 100% | 30% |
| deliverable_slack_message_content | 50% | 50% | 30% | 0% |
| floor_all_connectors | 100% | 90% | 90% | 30% |
| floor_read_source_ticket | 100% | 100% | 100% | 100% |
| floor_sql_investigation | 100% | 100% | 100% | 100% |
| floor_write_back | 100% | 90% | 100% | 30% |
| trace_all_connectors_and_trap | 80% | 50% | 40% | 0% |
| trace_auth_decline_pairing | 40% | 50% | 10% | 0% |