Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) on Split a shared house's tiered gas and electricity bills by days in the house, with unit conversion and leftover pennies
AClaude Haiku 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-23 · by @bobsiopener
- Estimated cost
- $0.0053
- Wall time
- 22s
- Human step-ins
- 0
- Outcome
- Pass
Compare with other models on this task →
Summary
Exact prompt
# Split a shared house's energy bills between housemates who came and went Inputs: `bills.csv` (bill_id, fuel, from, to, opening_read, closing_read), `housemates.csv` (name, move_in, move_out), `tariff.json` and `landlord-note.txt`. Amounts are in pounds. `tariff.json` and these rules are the record; where the landlord's note disagrees, the rules win. Rules: - A bill covers from to to, both days included. Units used = closing_read - opening_read; gas meters read in cubic metres, so gas kWh = units x volume_correction x calorific_value / kwh_divisor. - Each bill's first low_rate_kwh_per_day x (days in the bill) kWh are charged at low_rate and the rest at high_rate; add standing_per_day for every day; then add vat_pct percent. The bill total is rounded half-up to 0.01. - A housemate is in the house from move_in to move_out, both days included (blank move_out means still living there). Each bill total is split in proportion to each housemate's days in the house during that bill. - Each share is rounded half-up to 0.01; any penny difference between the shares and the bill total goes to the housemate with the most days in that bill (ties: first name alphabetically). - Write `out/split.json`: `{"kwh": {"<bill_id>": <kWh, 2 decimals>, ...}, "bill_totals": {"<bill_id>": <total>, ...}, "per_bill_shares": {"<bill_id>": {"<name>": <share>, ...}, ...}, "shares": {"<name>": <sum of their shares>, ...}}`. Per-bill shares list only housemates with at least one day in that bill; `shares` lists 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.
Work out each housemate's share of three months of gas and electricity bills. Gas meters read in cubic metres and need converting to kWh, the cheaper-rate allowance scales with each bill's length, standing charges and VAT are added, housemates moved in and out mid-bill with both days counted (one moved out on a bill's last day), and rounded shares must add back to the bill with the leftover penny going to a set person. A landlord's note gets the conversion, allowance, move-out and penny 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: gas_kwh tiers_and_vat occupancy_days pennies totals
Setup notes: Batch 31 case ta03-shared-house-energy-split (offline, synthetic inputs, seed 2906425906). Autonomous case: no stand-in user and no step-ins expected; any file the solver wrote to ask/ is counted as a measured intervention (0) and blocks a pass. Prompt = TASK.md verbatim. Hidden key kept outside the solver's work dir. Scored with the case's scorer (python3 score.py WORKDIR KEYDIR, absolute paths); pass = every check true and no ask/ files. Tools: Read/Write/Edit/Glob/Grep/Bash, web disabled. Single run, no retries. 1-min load average at scoring 0.27 (8 cores). Time-zone results depend on the system time-zone database: tzdata 2026c-0+deb13u1 (Python zoneinfo). Claude Code's own reported cost for claude-haiku-5.5: $0.005325. Score: 5/5. Cost is an ESTIMATE: measured tokens x pricing.json list price (Claude Pro subscription, nothing billed); input_tokens include cache reads and writes. Single run.
Estimated cost List-price estimate; not actually billed.
Metrics
- Estimated cost
- $0.0053
- Wall time
- 22s
- Human step-ins
- 0
- Turns
- 3
| Model | Input tok | Output tok | Cost |
|---|---|---|---|
| AClaude Haiku 5.5anthropic/claude-haiku-5.5effort default (no --effort flag) · thinking on | 64,918 | 4,553 | $0.0053 |
Reproduce this run
- Use the use case prompt on the use case page.
- Same setup: Claude Haiku 5.5 (anthropic/claude-haiku-5.5, effort default (no --effort flag) · thinking on) via
Claude Code 2.1.299 (headless -p), cloud. - It counts as a success only if it meets every success criterion: gas_kwh tiers_and_vat occupancy_days pennies totals
- Log model, tokens, wall time, human step-ins and the evidence log, then submit it as a reproduction. A different model? Submit a head-to-head rerun.
How runs are scored: Methodology.
Was this useful?
Automated checks
Trust 80% comes from these checks (how Trust % works).
- Run date is not in the future : passed
- Model is in the dated pricing table : passed
- Cost recomputed from tokens matches : passed
- Cached tokens are consistent : passed
- Output throughput is plausible : passed
- Autonomy matches step-ins : passed
- Step-ins do not exceed turns : passed
- Retries agree with first-try result : passed
- Evidence log SHA-256 matches : passed
- Evidence log is substantial : passed
Evidence log
[
{
"duration_api_ms": 21022,
"stop_reason": "end_turn",
"session_id": "[REDACTED]",
"total_cost_usd": 0.00532516,
"usage": {
"input_tokens": 6,
"cache_creation_input_tokens": 12626,
"cache_read_input_tokens": 52286,
"output_tokens": 4553,
"output_tokens_details": {
"thinking_tokens": 2520
},
"server_tool_use": {
"web_search_requests": 0,
"web_fetch_requests": 0
},
"service_tier": "standard",
"cache_creation": {
"ephemeral_1h_input_tokens": 12626,
"ephemeral_5m_input_tokens": 0
},
"inference_geo": "global",
"iterations": [
{
"input_tokens": 2,
"output_tokens": 1031,
"cache_read_input_tokens": 20402,
"cache_creation_input_tokens": 4763,
"cache_creation": {
"ephemeral_5m_input_tokens": 0,
"ephemeral_1h_input_tokens": 4763
},
"type": "message"
}
],
"speed": "standard",
"fallback_credit": null
},
"modelUsage": {
"claude-haiku-5-5": {
"inputTokens": 6,
"outputTokens": 4553,
"cacheReadInputTokens": 52286,
"cacheCreationInputTokens": 12626,
"webSearchRequests": 0,
"costUSD": 0.00532516,
"contextWindow": 1000000,
"maxOutputTokens": 128000,
"thinkingTokens": 2520,
"canonicalModel": "claude-haiku-5-5",
"provider": "firstParty",
"costBasis": "list"
}
},
"permission_denials": [],
"terminal_reason": "completed",
"fast_mode_state": "off",
"fast_mode_disabled_reason": "sdk_opt_in_required",
"subagent_stats": {
"spawned": 0,
"requested": {
"background": 0,
"foreground": 0,
"unset": 0
},
"started_in_background": 0,
"max_depth": 0,
"spawned_by_subagents": 0,
"completed": 0,
"failed": 0,
"killed": {
"parent": 0,
"user": 0,
"system": 0
},
"refused": {
"depth_limit": 0,
"concurrency_limit": 0,
"budget": 0
},
"by_type": {}
},
"safety_stops": 0,
"is_error": false,
"num_turns": 3,
"subtype": "success",
"api_error_status": null,
"result": "The split is written to `out/split.json`. I followed `tariff.json` and the rules wherever the landlord's note disagreed with them:\n\n- **Gas:** the meter reads in cubic metres, so I applied the volume correction and calorific value. The note says the meter already reads kWh.\n- **Cheap-rate allowance:** charged per day of each bill (6 kWh/day electricity, 30 kWh/day gas), not per month.\n- **Move-out day:** counted as a day in the house, so Ade's 4 August and Fay's 12 August are both included. The note says nobody pays for their move-out day.\n\nResults:\n\n| Bill | kWh | Total (\u00a3) |\n|---|---|---|\n| ELEC-1 | 208.00 | 65.52 |\n| GAS-1 | 1370.22 | 109.71 |\n| ELEC-2 | 298.00 | 93.87 |\n| GAS-2 | 1558.08 | 124.35 |\n| ELEC-3 | 230.00 | 72.45 |\n| GAS-3 | 1314.97 | 104.20 |\n\nOverall shares: Dot \u00a3215.46, Cal \u00a3171.01, Ade \u00a3137.52, Fay \u00a346.11. Each bill's shares sum to its total.\n\nPenny adjustments went to the housemate with the most days in that bill, with ties broken alphabetically. ELEC-2 is a three-way tie (Ade, Cal, Dot at 33 days), so Ade took the \u2212\u00a30.01. ELEC-3 is a tie between Cal and Dot at 30 days, so Cal took the +\u00a30.01.\n\nI computed everything with exact decimal arithmetic in a script at `/tmp/split_calc.py`, outside `out/`. I did not write anything under `ask/`.",
"ttft_ms": 1750,
"type": "result",
"duration_ms": 21382,
"uuid": "[REDACTED]",
"ttft_stream_ms": 1562,
"time_to_request_ms": 262,
"first_content_frame_ms": 1562,
"queued_turn_count": 0,
"result_index": 0
}
]