Skip to content

Share out a restaurant's weekly tip pool by day, with a card-fee deduction, a bar tip-out and points by role

Exact prompt

# Share out last week's tip pool Inputs: `shifts.csv` (who worked, in which role, for how many hours, per day), `tips.csv` (tips taken per day, cash or card) and `tip-policy.json`. Work out each person's tips for the week 2026-10-12 to 2026-10-18. - The pool is settled separately for each day, using only that day's tips and shifts. - Card tips are reduced by the card processing fee (`card_fee_pct`) before anything else; cash tips are not. - From each day's pool, `bar_share_pct` percent goes to the bartenders who worked that day, split by their hours that day. - The rest of the day's pool is split among servers and bussers in proportion to hours x role points (`points` in the policy). Managers take no tips. - Work exactly; round only the reported figures, to 2 decimals half-up. Write `out/tips.json`: `{"tips": {"<staff_id>": <week total>, ...}, "bar_total": <week total paid to bartenders>, "fees_deducted": <card fees for the week>}`. List everyone who received tips. ## 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.

Split a week of a restaurant's cash and card tips among staff from the shift log. The pool is settled day by day, card tips lose the processing fee first, the bar takes a fixed share split by bartender hours, the rest is shared by hours weighted by role points, and managers get nothing. Pooling the whole week, skipping the card fee or paying managers each changes someone's tips. 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: daily_pool card_fee bar_tipout points_no_managers

Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0044, 15s).

Run this on another model

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.ⓘ

Runs for this use case
ModelCostTimeStep-insⓘAutonomyOutcomeTrustⓘDateCompare
AClaude Opus 5.5default · thinking on$0.1321s0Fully autonomouspassTrust 80%2026-10-21
AClaude Haiku 5.5default · thinking on$0.004415s0Fully autonomouspassTrust 80%2026-10-21
AClaude Sonnet 5.5default · thinking on$0.0718s0Fully autonomouspassTrust 80%2026-10-21

Embed this result

A live badge for this use case. It links back here and updates as new runs are published.

UseCaseRank badge for Share out a restaurant's weekly tip pool by day, with a card-fee deduction, a bar tip-out and points by role

Get new runs and ranking changes weekly

Which models got better or cheaper on real tasks like this one.