Dropped incremental-auth phantom-hold reconcileCards & 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.
100%
GPT-5.5
91%
Claude Opus 4.8
86%
Gemini 3.1 Pro
23%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
6m
Gemini 3.5 Flash
4m
Gemini 3.1 Pro
2m
GPT-5.5
2m
Claude Opus 4.8
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
37
Gemini 3.5 Flash
23
Gemini 3.1 Pro
17
Claude Opus 4.8
12
GPT-5.5
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 (17) | ||||
| cp_affected_trace_count | 100% | 80% | 80% | 0% |
| cp_repeat_offender_accounts | 100% | 80% | 80% | 0% |
| cp_top_affected_account_id | 100% | 100% | 100% | 40% |
| cp_traces_with_incremental_auth | 100% | 100% | 100% | 80% |
| deliverable_jira_ticket_filed | 100% | 100% | 60% | 20% |
| deliverable_slack_channel_exists | 100% | 100% | 100% | 20% |
| deliverable_slack_message_content | 100% | 80% | 80% | 0% |
| cp_affected_account_count | 100% | 80% | 80% | 0% |
| cp_dropped_increment_dollars_cents | 100% | 80% | 80% | 0% |
| cp_top_account_dropped_cents | 100% | 80% | 80% | 0% |
| cp_top_account_trace_count | 100% | 100% | 100% | 40% |
| floor_all_connectors | 50% | 0% | 40% | 20% |
| floor_read_source_ticket | 50% | 0% | 40% | 100% |
| floor_sql_investigation | 100% | 100% | 100% | 100% |
| floor_write_back | 100% | 100% | 100% | 20% |
| trace_all_connectors_and_trap | 100% | 80% | 30% | 0% |
| trace_card_events_hold_reconstruct | 100% | 100% | 90% | 80% |