VIP ACH Limit Compliance: non-VIP weekly PaySomeone ACH auditPayments & Transfers
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.
62%
GPT-5.5
58%
Gemini 3.1 Pro
43%
Claude Opus 4.8
31%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
17m
Gemini 3.5 Flash
4m
Gemini 3.1 Pro
3m
GPT-5.5
2m
Claude Opus 4.8
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
60
Gemini 3.5 Flash
40
Gemini 3.1 Pro
32
GPT-5.5
22
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 (15) | ||||
| ach_outbound_scope | 50% | 50% | 60% | 40% |
| cp1_vip_eligible_accounts_count | 20% | 0% | 30% | 0% |
| cp3_non_vip_weekly_limit_usd | 100% | 100% | 100% | 90% |
| cp4_pager_duty_alert_threshold_usd | 100% | 100% | 100% | 90% |
| cp5_total_ach_debit_transfers | 40% | 40% | 60% | 20% |
| cp7_non_vip_accounts_exceeding_weekly_limit | 20% | 10% | 30% | 0% |
| jira_pd177_vip_criteria | 100% | 90% | 100% | 30% |
| jira_pd346_comment_policy | 60% | 0% | 0% | 0% |
| vip_derivation_full_tables | 70% | 0% | 40% | 10% |
| cp2_vip_eligible_pct | 80% | 0% | 40% | 0% |
| cp6_total_ach_debit_volume_usd | 40% | 40% | 60% | 20% |
| cp8_compliance_violation_count | 20% | 20% | 30% | 0% |
| process_jira_floor | 70% | 100% | 100% | 100% |
| process_sql_floor | 100% | 100% | 100% | 100% |
| weekly_grouping_logic | 100% | 90% | 80% | 0% |
Failure modes on this task
40 tagged verifier failures, by the stage they were assigned to. Higher is worse.
| Failure stage | GPT-5.5 | Claude Opus 4.8 | Gemini 3.1 Pro | Gemini 3.5 Flash |
|---|---|---|---|---|
| Answer3 modes | 10 | 10 | 10 | 10 |