Invoice a nursery's families for a month with weekly funded hours, a term-start age rule and a sibling discount
Exact prompt
# Invoice a nursery's families for October 2026 Inputs: `children.csv` (child id, family id, date of birth), `attendance.csv` (hours each child attended each day in October 2026) and `fees.json`. Invoice rules: - Funded hours: a child who was 3 or older on 2026-09-01 gets up to `funded_hours_per_week` hours free in each Monday-to-Sunday week (the part-week 1-4 October counts as a week); unused funded hours don't carry to another week. - Charged hours = attended hours minus funded hours used; hours fee = charged hours x `hourly_rate`. - Sibling discount: `sibling_discount_pct` off the hours fee (not meals) of every child in a family except the oldest child who attended in October. - Meals: `meal_per_day` for each day attended. - Each child's total = hours fee after discount + meals, rounded half-up to the cent. A family's total is the sum of its children's rounded totals. Write `out/invoices.json`: `{"funded_hours_used": {"<child_id>": <hours>, ...}, "child_totals": {"<child_id>": <total>, ...}, "family_totals": {"<family_id>": <total>, ...}}` for every child and family that attended in October. ## How this run works - Work on your own: nobody will answer questions in this task, and writing to `ask/` counts against the run. - Write only under `out/`. Create `out/` if it is missing. Inputs are synthetic; don't fetch anything from the network. - Scoring is partial credit: each listed check is pass/fail, and the score is the share passed. Numbers are compared to within 0.005 unless the task says otherwise.
Bill each family at a day nursery for October from daily attendance. Children aged 3 by the start of term get a weekly allowance of free hours that does not roll over, younger siblings get a discount on their hours fee but not their meals, and totals are rounded per child. Pooling the free hours over the month, judging age on the invoice date or skipping the sibling discount each change the invoices. Original benchmark task written for usecaserank.com (not based on a specific public post). Headless, offline run on a reproducible fixture. Each run's notes say how its cost was measured (estimates are marked).
Success criteria: funded_weekly eligibility_date sibling_discount child_totals
Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0052, 19s).
All runs (3)
100% pass rate across 3 eligible runs: 3 pass, 0 partial, 0 fail. Partial and failed runs are listed below with everything else.ⓘ
| Model | Cost | Time | Step-insⓘ | Autonomy | Outcome | Trustⓘ | Date | Compare |
|---|---|---|---|---|---|---|---|---|
| AClaude Opus 5.5default · thinking on | $0.15 | 30s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-22 | |
| AClaude Haiku 5.5default · thinking on | $0.0052 | 19s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-22 | |
| AClaude Sonnet 5.5default · thinking on | $0.06 | 14s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-22 |
Embed this result
A live badge for this use case. It links back here and updates as new runs are published.
Get new runs and ranking changes weekly
Which models got better or cheaper on real tasks like this one.