| { | |
| "cases": 75, | |
| "choices_matched": 75, | |
| "max_probability_error": 0.018875091555677126, | |
| "reference_peak_rss_bytes": 1244889088, | |
| "limitations": [ | |
| "Synthetic conversion-fidelity tests, not a quality or calibration benchmark.", | |
| "Text-only artifact; images and video are rejected.", | |
| "No iPhone or accelerator-placement verification." | |
| ], | |
| "business": { | |
| "cases": 48, | |
| "correct": 24, | |
| "reference_correct": 24, | |
| "confusion": { | |
| "interrupt": { | |
| "interrupt": 2, | |
| "discard": 6, | |
| "enqueue": 3 | |
| }, | |
| "discard": { | |
| "interrupt": 0, | |
| "discard": 9, | |
| "enqueue": 5 | |
| }, | |
| "enqueue": { | |
| "interrupt": 0, | |
| "discard": 10, | |
| "enqueue": 13 | |
| } | |
| }, | |
| "false_discard": 16, | |
| "false_interrupt": 0, | |
| "missed_interrupt": 9 | |
| } | |
| } | |