Onboarding DOB timezone shift auditCustomer & Account Analytics
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.
80%
GPT-5.5
56%
Claude Opus 4.8
31%
Gemini 3.1 Pro
1%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
105m
Gemini 3.1 Pro
8m
Gemini 3.5 Flash
7m
Claude Opus 4.8
5m
GPT-5.5
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
125
Gemini 3.1 Pro
61
Claude Opus 4.8
60
GPT-5.5
57
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_affected_accounts | 60% | 0% | 40% | 0% |
| answer_affected_population | 60% | 0% | 80% | 10% |
| answer_shift_direction_split | 40% | 0% | 30% | 0% |
| deliverable_jira_ticket_filed | 100% | 100% | 10% | 0% |
| deliverable_slack_channel_exists | 100% | 100% | 20% | 0% |
| deliverable_slack_message_content | 100% | 90% | 20% | 0% |
| deliverable_slack_pin | 100% | 100% | 20% | 0% |
| floor_all_connectors | 100% | 100% | 20% | 0% |
| floor_read_source_ticket | 100% | 100% | 100% | 80% |
| floor_sql_investigation | 100% | 100% | 100% | 100% |
| floor_write_back | 100% | 100% | 20% | 0% |
| trace_all_connectors_and_trap | 60% | 0% | 20% | 0% |
| trace_dob_shift_method | 60% | 0% | 10% | 0% |