Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) on Work out late-payment interest and compensation from part payments, credit notes and a base-rate change

AClaude Haiku 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-23 · by @bobsiopener

Trust 80%Fully autonomoussuccess · first tryCompare

Estimated cost: $0.0079
Wall time: 38s
Human step-ins: 0
Outcome: Pass

Compare with other models on this task →

Exact prompt

# Work out a design studio's late-payment interest and compensation Inputs: `invoices.csv` (inv_id, client, issued, due, amount), `credits.csv` (inv_id, amount), `payments.csv` (payment_id, client, paid_on, amount), `base-rates.csv` (from_date, rate), `terms.json` and `bookkeeper-note.txt`. Amounts are in pounds. `terms.json` and these rules are the record; where the bookkeeper's note disagrees, the rules win. Rules: - Clients match ignoring letter case and surrounding spaces; report them in upper case. A credit note reduces its invoice's amount from the start. - Payments are applied in paid_on order (ties by payment_id), each to the client's invoices with the earliest due date first (ties by the invoice number), filling one before the next; anything left over is unapplied. - Interest accrues for each day after the due date up to and including as_of on the amount still unpaid at the end of that day, at (statutory_pct + the base rate in force on that day) percent a year, divided by 365; each invoice's interest is rounded half-up to 0.01 once, at the end. - An invoice still unpaid at as_of, or paid in full only after its due date, earns one compensation fee from the first terms.json band whose `below` is more than its amount after credit notes (`below` null is the top band); an invoice paid in full on or before its due date earns none. - Write `out/late-payment.json`: `{"balances": {"<inv_id>": <unpaid at as_of>, ...}, "unapplied": {"<CLIENT>": <amount>, ...}, "interest": {"<inv_id>": <interest>, ...}, "compensation": {"<inv_id>": <fee>, ...}, "client_totals": {"<CLIENT>": <balances + interest + compensation>, ...}}`. `balances` and `client_totals` cover every invoice and client; the other maps list only amounts above 0. ## 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.

Allocate a studio's client payments to invoices and work out statutory late-payment interest and fixed compensation. Payments clear the earliest due invoice first with invoice numbers that cross 99 to 100, credit notes cut one invoice below a compensation band edge, the base rate changes part-way through the overdue period, one invoice is paid exactly on its due date, and client names come in mixed case. A bookkeeper's note gets the rate, allocation and banding 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: allocation interest compensation client_totals

Setup notes: Batch 30 case o03-late-payment-interest (offline, synthetic inputs, seed 3047022779). 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 0.38 (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: $0.007913. Score: 4/4. 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: $0.0079
Wall time: 38s
Human step-ins: 0
Turns: 4

Model usage
Model	Input tok	Output tok	Cost
AClaude Haiku 5.5anthropic/claude-haiku-5.5effort default (no --effort flag) · thinking on	90,978	7,884	$0.0079

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: allocation interest compensation client_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?

⇄ Rerun on another model

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": 36551,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.007912830000000001,
  "usage": {
   "input_tokens": 8,
   "cache_creation_input_tokens": 16107,
   "cache_read_input_tokens": 74863,
   "output_tokens": 7884,
   "output_tokens_details": {
    "thinking_tokens": 5267
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 16107,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 2005,
     "cache_read_input_tokens": 22004,
     "cache_creation_input_tokens": 6642,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 6642
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-haiku-5-5": {
    "inputTokens": 8,
    "outputTokens": 7884,
    "cacheReadInputTokens": 74863,
    "cacheCreationInputTokens": 16107,
    "webSearchRequests": 0,
    "costUSD": 0.007912830000000001,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 5267,
    "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": 4,
  "subtype": "success",
  "api_error_status": null,
  "result": "I wrote `out/late-payment.json`. The interest and fee rules follow `TASK.md`, not the bookkeeper's note, which I overrode in three places.\n\n**Results (as of 2026-11-30):**\n\n| Client | Balances | Interest | Compensation | Total |\n|---|---|---|---|---|\n| BRAMBLE CAFE | 0.00 | 4.75 | 110.00 | 114.75 |\n| NORTHGATE DENTAL | 103.71 | 16.02 | 120.00 | 239.73 |\n| PELLOW & SONS | 3,689.10 | 92.77 | 110.00 | 3,891.87 |\n\n- **Unapplied:** BRAMBLE CAFE has 1,280.88 left over after its payments.\n- **Still open:** INV-108 (3,689.10) and INV-112 (103.71).\n\n**Where I departed from the bookkeeper's note:**\n- **Base rate:** interest uses the rate in force on each day (4.25% until 2026-10-21, 3.75% from 2026-10-22), not the rate from the due date.\n- **Payment order:** payments clear the earliest-due invoice first, not the newest.\n- **Compensation basis:** the fee band uses the amount after credit notes, not the gross invoice total.\n\n**Method:** I stepped through each day from the earliest issue date to as_of. Payments for a day are applied first, then that day's interest accrues on the end-of-day balance. Interest is summed exactly as fractions and rounded half-up to 0.01 once per invoice. I spot-checked INV-108's interest by hand and it matches to the cent.\n\nI ran the calculation inline without saving a script, so nothing was written outside `out/`. I did not commit anything.",
  "ttft_ms": 1282,
  "type": "result",
  "duration_ms": 36950,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 1084,
  "time_to_request_ms": 249,
  "first_content_frame_ms": 1085,
  "queued_turn_count": 0,
  "result_index": 0
 }
]