bench_runs

{“what”:“Unique numeric identifier for the run; auto-incremented.”,“why”:“Without it, rows cannot be uniquely identified or updated.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.824078299+00:00”}

{“what”:“Name or path of the benchmark fixture used for this run.”,“why”:“Without it, results cannot be attributed to a specific test case.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.877553257+00:00”}

{“what”:“Count of tokens in the input text for the run.”,“why”:“Without it, token usage and compression ratio cannot be computed.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.888543107+00:00”}

{“what”:“Count of tokens in the generated summary.”,“why”:“Without it, summary length and compression ratio are unknown.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.902177803+00:00”}

{“what”:“Ratio of summary tokens to input tokens; unitless.”,“why”:“Without it, runs cannot be sorted by compression efficiency without recomputing.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.921412877+00:00”}

{“what”:“Time taken to seal the run, in milliseconds.”,“why”:“Without it, performance of the sealing step cannot be measured.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.939488569+00:00”}

{“what”:“Percentage of facts from the input retained in the summary, 0-100.”,“why”:“Without it, factual completeness of summaries is unknown.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:14.990945145+00:00”}

{“what”:“Judge-assigned quality score for the summary.”,“why”:“Without it, summary quality cannot be compared across runs.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:15.016180733+00:00”}

{“what”:“Timestamp when the run was recorded, in UTC.”,“why”:“Without it, the chronological order of runs is unknown.”,“by”:“fleet-annotator-worker”,“at”:“2026-09-23T00:17:15.032678482+00:00”}