# Attribution pilot — speakers-smoke

Lane **diarized** · sample `video:shondo/OWuZuOrbQ10` · 1 video(s) · model `qwen2.5:7b` @ numCtx 8192 (maxCues 600) · 2026-08-12T18:18:55.516Z

Outcomes: attributed 1

## Diarized lane — the claim under test is ONE call per video

| video | calls | clusters offered | clusters named | confidences |
| --- | ---: | ---: | ---: | --- |
| shondo/OWuZuOrbQ10 | 1 | 3 | 3 | 0.90, 0.50, 0.80 |

**Calls per video: 1.00** (1 call(s) over 1 video(s)). The claim holds iff this is 1.00.

Warnings by code: none

## Cost

Measured over the 1 video(s) this run actually attributed (1/1 chunk(s) ok, 1 call(s), 1 with engine timing). Sample density **1.41 chunks/audio-hour** — the corpus census reads 2.46.

| quantity | value |
| --- | ---: |
| s/chunk, wall (video wall ÷ chunks) | 14.4 |
| s/chunk, per-call wall | 14.4 |
| s/chunk, engine | 14.3 |
| s/chunk, engine excl. model load | 4.7 |
| mean model-load s/call | 0.40 |
| decode tokens/s | 42.7 |
| digest chapters baseline, s/chunk engine | 11.2 |

## Box

Start: `{"loadavg":[13.83,16.17,15.04],"freeMemGb":2.61,"totalMemGb":15.53}`

End: `{"loadavg":[12.17,15.69,14.9],"freeMemGb":2.36,"totalMemGb":15.53}`

## Units

- The unit of text-only work is the **chunk** (one model call). Seconds-per-
  audio-hour is not a unit here: chunk density varies 4× across this corpus, and
  pricing in it is the error behind a retracted throughput headline. This report
  does not print one.
- The unit of diarized work is **calls per video**.
- Wall and engine are different quantities and are never substituted for each
  other. Wall is the realistic ceiling on this shared box; engine excl. load is
  the floor.
- Every rate carries its denominator. At n ≈ 8 videos a bare percentage would be
  a lie of precision.