Post-closure force-capture exposureCards & 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
88%
Claude Opus 4.8
84%
Gemini 3.1 Pro
36%
Gemini 3.5 Flash
Extended metrics
Median Time
Wall-clock minutes, median across trials. N = 10 trials per model.
16m
Gemini 3.5 Flash
3m
GPT-5.5
2m
Gemini 3.1 Pro
1m
Claude Opus 4.8
Tool Calls
Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.
39
Gemini 3.5 Flash
19
Gemini 3.1 Pro
16
GPT-5.5
15
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 (7) | ||||
| cp1_card_processor | 100% | 100% | 90% | 90% |
| cp2_post_closure_capture_usd | 100% | 80% | 80% | 0% |
| cp3_post_closure_capture_count | 100% | 80% | 80% | 0% |
| cp4_can_prevent | 100% | 100% | 90% | 90% |
| cp5_accounts_affected | 100% | 80% | 80% | 0% |
| process_floor | 100% | 100% | 100% | 100% |
| rubric_auth_vs_clear | 100% | 80% | 90% | 0% |