Split a shared house's utility bills by days of occupancy, to the cent
Exact prompt
# Split a shared house's utility bills by days of occupancy Inputs: `housemates.json` (each person's move-in and move-out dates; null means still living there), `bills.json` (each bill's amount, service period and who paid it). Rules: - A person occupies the house on every date from move_in to move_out, both inclusive. - For each bill, a person's days = the dates inside the bill's service period (both ends inclusive) that they occupy the house. - Share = bill amount x person's days / sum of all people's days for that bill, rounded half-up to the cent. - If the rounded shares don't add up to the bill amount, add or remove the difference one cent at a time, starting with the person with the most days on that bill (ties: alphabetical by name), cycling in that order. - People with 0 days on a bill owe nothing for it and are left out of that bill's shares. - net_usd = what the person paid minus the total of their shares (positive means they are owed money). Write `out/split.json`: `{"days": {"<bill_id>": {"<name>": <int>, ...}}, "shares": {"<bill_id>": {"<name>": <usd>, ...}}, "owed_usd": {"<name>": <usd>}, "net_usd": {"<name>": <usd>}}`. `owed_usd` and `net_usd` cover every housemate. ## 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.
Prorate six utility bills across 4-6 housemates who moved in and out on different dates, then work out what each owes and their net balance. Shares are rounded half-up to the cent, and a stated cent-by-cent rule makes each bill's shares add up exactly. Draws whose shares fall within 1e-4 of a half-cent tie are rejected. 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: occupancy_days prorated_shares shares_sum_to_bill zero_day_people_excluded owed_totals net_balances
Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0048, 20s).
All runs (4)
100% pass rate across 4 eligible runs: 4 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 |
|---|---|---|---|---|---|---|---|---|
| DDeepSeek V4.1 Flash (Venice)thinking on | $0.0052 | 17s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-28 | |
| AClaude Opus 5.5default · thinking on | $0.11 | 19s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-20 | |
| AClaude Haiku 5.5default · thinking on | $0.0048 | 20s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-20 | |
| AClaude Sonnet 5.5default · thinking on | $0.05 | 10s | 0 | Fully autonomous | pass | Trust 80% | 2026-10-20 |
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.