Mastercard qualifying signature volume in CY2024Cards & 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.
81%
GPT-5.5
77%
Gemini 3.1 Pro
73%
Claude Opus 4.8
19%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
4m
GPT-5.5
4m
Gemini 3.1 Pro
3m
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.
28
Gemini 3.1 Pro
28
Gemini 3.5 Flash
24
GPT-5.5
23
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 (24) | ||||
| answer_bo25_pulse_trend_verdict | 80% | 100% | 60% | 10% |
| answer_bo8_claim_verdict | 60% | 80% | 20% | 10% |
| deliverable_jira_comment_on_bo8 | 100% | 100% | 100% | 0% |
| deliverable_slack_channel_exists | 100% | 100% | 100% | 0% |
| deliverable_slack_message_content | 80% | 60% | 30% | 0% |
| deliverable_slack_message_pinned | 100% | 100% | 100% | 0% |
| gold_all_network_signature_share_pct | 60% | 40% | 80% | 10% |
| gold_card_transaction_volume_2024_usd | 90% | 50% | 100% | 40% |
| gold_documented_blended_take_rate_pct | 100% | 100% | 100% | 80% |
| gold_implied_interchange_at_blended_rate_usd | 80% | 80% | 50% | 30% |
| gold_non_transaction_volume_2024_usd | 70% | 50% | 100% | 50% |
| gold_pulse_share_h1_2024_pct | 90% | 40% | 100% | 10% |
| gold_pulse_share_h2_2024_pct | 90% | 40% | 100% | 10% |
| gold_qualifying_distinct_transaction_count_2024 | 40% | 90% | 70% | 0% |
| gold_qualifying_share_of_transactions_pct | 70% | 50% | 70% | 20% |
| gold_qualifying_signature_volume_2024_usd | 80% | 80% | 50% | 40% |
| floor_all_connectors | 100% | 100% | 100% | 0% |
| floor_read_jira_policy | 100% | 100% | 100% | 100% |
| floor_sql_investigation | 100% | 100% | 100% | 100% |
| gold_ledger_approved_volume_2024_usd | 70% | 90% | 100% | 40% |
| gold_qualifying_events_with_amount_2024 | 80% | 80% | 50% | 0% |
| gold_qualifying_share_of_ledger_pct | 50% | 70% | 50% | 20% |
| trace_non_transaction_discovery | 70% | 90% | 100% | 20% |
| trace_pulse_time_series | 100% | 100% | 100% | 10% |
Failure modes on this task
35 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 |
|---|---|---|---|---|
| Answer4 modes | 10 | 6 | 9 | 10 |