Skip to content

Total an apiary group's honey harvest by site and hive after hive moves, corrections and a moisture rule

Exact prompt

# Total an apiary group's honey harvest by site and hive after hive moves, corrections and a moisture rule Inputs: `sites.csv` (site_id, name), `hives.csv` (hive_id, start_site), `moves.csv` (hive_id, date, to_site) and `harvests.csv` (harvest_id, date, hive_id, frames, gross_kg, moisture_pct, corrects). The season's ledger is synthetic: every site, hive and beekeeper is invented, and the tare and moisture figures below are this group's own house rules. These rules are the group's; where `helper-note.txt` disagrees, the rules win. Rules: - Corrections: a harvest row whose corrects names another harvest_id replaces that row completely; the corrected row is ignored and the correction's own date, hive and figures are used. A blank corrects means an ordinary row. - Sites: a harvest counts at the site its hive stood at on the harvest date: the hive's start_site, changed by every move of that hive dated on or before the harvest date (a move counts from its own date). - Net honey: gross_kg minus 0.42 kg of tare for each frame, in kg. - Feed honey: a harvest whose moisture_pct is above 18.5 is set aside as feed honey; at 18.5 or below its net honey is sellable. - Top hive and jars: the top hive has the most sellable honey (on a tie, the lower hive_id); jars is the number of whole 340 g jars the total sellable honey fills. - Write `out/harvest.json`: `{"sellable_kg_by_site": {"<site_id>": <kg>, ...}, "feed_kg_by_site": {"<site_id>": <kg>, ...}, "sellable_kg_by_hive": {"<hive_id>": <kg>, ...}, "top_hive": "<hive_id>", "jars": <whole number>}`. - Roll-ups: Sites in sites.csv with no sellable honey are left out. Sites in sites.csv with no feed honey are left out. Hives in hives.csv with no sellable honey are left out. ## 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.

Work out a synthetic beekeeping group's season from a harvest ledger: net honey is gross weight minus a stated tare per frame, each harvest counts at the site the hive stood at on that date (hives are moved between sites, and a move counts from its own date), correction rows replace the rows they correct, and honey over a stated moisture limit is set aside as feed honey. Report sellable and feed honey per site, sellable honey per hive, the top hive and the jar count. The prompt says sites and hives with nothing to report are left out, not listed at 0. A helper's note gets the tare, move and roll-up rules wrong. 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: sellable_by_site feed_by_site sellable_by_hive top_hive jars

Best model for this task so far: Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) (success, $0.0042, 16s).

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$0.1221s0Fully autonomouspassTrust 80%2026-10-26
AClaude Haiku 5.5default · thinking on$0.004216s0Fully autonomouspassTrust 80%2026-10-26
AClaude Sonnet 5.5default · thinking on$0.0511s0Fully autonomouspassTrust 80%2026-10-26

Embed this result

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

UseCaseRank badge for Total an apiary group's honey harvest by site and hive after hive moves, corrections and a moisture rule

Get new runs and ranking changes weekly

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