Skip to content

Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) on Build a step-by-step interactive explainer web app that teaches a beginner how to choose an LED's series resistor

AClaude Haiku 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-22 · by @bobsiopener

Trust 80%Fully autonomoussuccess · first try
Estimated cost
$0.01
Wall time
1m
Human step-ins
0
Outcome
Pass

Compare with other models on this task →

Summary

Exact prompt

# Build a step-by-step explainer web app: choosing a series resistor for an LED Make a small interactive web app that teaches a beginner one hardware topic, one step at a time: how to pick the series resistor for an LED. `spec.json` gives the page title, the steps (each with a title and a fact sentence), and the standard resistor series to use (`series_base`, one decade of values, repeated across `decades`). Look, layout and wording beyond the spec are up to you; **looks are not scored**. The page is scored in a headless browser (800x600 window, network blocked) by objective checks only. Write one file, `out/explainer.html`, that works opened straight from disk. ## Steps - Show one step at a time. On load, step 1 is shown. Put the current step's title in `#step-title`, its text in `#step-body` and a counter in `#step-count`. - Buttons `#prev` and `#next` move one step back or forward. - Every step title must appear in `#step-title` in the order given in spec.json, and `#step-count` reads "Step <i> of <N>". - `#prev` is disabled (the `disabled` property) on step 1 and `#next` is disabled on the last step; neither moves past the ends. - `#step-body` for each step contains that step's fact sentence from spec.json exactly as written (extra text around it is fine). ## Calculator Inputs `#in-supply` (supply volts), `#in-vf` (LED forward volts), `#in-ma` (LED current in mA) and `#in-count` (LEDs in series, a whole number); outputs `#out-ohms`, `#out-standard`, `#out-mw` and `#out-error`. The calculator is visible on every step and recalculates on each `input` or `change` event (a `#calc` button is optional; it is clicked too if present). - For supply V, forward voltage F, current I mA and count n: R = (V - n x F) / (I / 1000). `#out-ohms` shows R rounded half-up to a whole ohm. - `#out-standard` shows the smallest standard series value that is at or above the exact R, never a smaller one. - `#out-mw` shows the resistor's power, (V - n x F) x I, in mW rounded half-up to a whole number. - Show numbers as plain digits (for example `1200`, not `1.2k`); a unit after the number is fine. - If any input is missing or not positive, or V is not greater than n x F, `#out-error` shows a non-empty message and `#out-ohms`, `#out-standard` and `#out-mw` are empty. Otherwise `#out-error` is empty. ## Checks (partial credit = share passed) 1. `self_contained`: `out/explainer.html` exists, is at most 150,000 bytes, references no http(s) or protocol-relative URL, and requests nothing from the network when opened. 2. `required_elements`: every element id named above is present, and the `<title>` contains spec.json's `title`. 3. `steps_in_order`: Every step title must appear in `#step-title` in the order given in spec.json, and `#step-count` reads "Step <i> of <N>". 4. `navigation_limits`: `#prev` is disabled (the `disabled` property) on step 1 and `#next` is disabled on the last step; neither moves past the ends. 5. `step_facts`: `#step-body` for each step contains that step's fact sentence from spec.json exactly as written (extra text around it is fine). 6. `calculator_values`: on hidden test inputs, `#out-ohms`, `#out-standard` and `#out-mw` all match the rules above. 7. `invalid_input`: If any input is missing or not positive, or V is not greater than n x F, `#out-error` shows a non-empty message and `#out-ohms`, `#out-standard` and `#out-mw` are empty. No build step and no other files are scored. ## 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.

Build a one-file web app that walks a beginner through choosing an LED's series resistor, one step at a time, with a live calculator. A headless browser clicks through the steps and feeds hidden inputs to the calculator, checking step order, the button limits at both ends, the given facts, the calculated values (including rounding up to a standard resistor value) and the error state. Looks are not scored. 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: self_contained required_elements steps_in_order navigation_limits step_facts calculator_values invalid_input

Setup notes: Batch 26 case b01-led-resistor-explainer (offline, synthetic inputs, seed 143914568). 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 0.24 (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: $0.013398. Score: 7/7. 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
$0.01
Wall time
1m
Human step-ins
0
Turns
9
Model usage
ModelInput tokOutput tokCost
AClaude Haiku 5.5anthropic/claude-haiku-5.5effort default (no --effort flag) · thinking on266,08112,533$0.01

Reproduce this run

  1. Use the use case prompt on the use case page.
  2. 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.
  3. It counts as a success only if it meets every success criterion: self_contained required_elements steps_in_order navigation_limits step_facts calculator_values invalid_input
  4. 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).

Evidence log
[
 {
  "duration_api_ms": 65923,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.013397919999999999,
  "usage": {
   "input_tokens": 18,
   "cache_creation_input_tokens": 23521,
   "cache_read_input_tokens": 242542,
   "output_tokens": 12533,
   "output_tokens_details": {
    "thinking_tokens": 6825
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 23521,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 663,
     "cache_read_input_tokens": 35228,
     "cache_creation_input_tokens": 832,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 832
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-haiku-5-5": {
    "inputTokens": 18,
    "outputTokens": 12533,
    "cacheReadInputTokens": 242542,
    "cacheCreationInputTokens": 23521,
    "webSearchRequests": 0,
    "costUSD": 0.013397919999999999,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 6825,
    "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": 9,
  "subtype": "success",
  "api_error_status": null,
  "result": "The page is at `out/explainer.html`, about 7.5 KB, and it passes every check I could run without a real browser.\n\n- **Steps:** all 7 titles and fact sentences are embedded from `spec.json` by script, not retyped. The page starts on step 1 with `#prev` disabled, and `#next` is disabled on step 7.\n- **Calculator:** I ran the formulas in Node against a stubbed DOM. For example, 9 V, 2.4 V, 13 mA, 2 LEDs gives 323 ohms, standard 330, 55 mW. Invalid inputs (empty, non-positive, non-whole LED count, supply not above n \u00d7 forward voltage) give an error message and empty results.\n- **Self-contained:** no external URLs or network requests, and no `http` strings anywhere in the file.\n\nI did not open it in the headless browser, so the DOM and event wiring are untested beyond the stub. Listed checks I'd expect to pass: 1\u20137.\n\nChoices I made that the spec left open:\n- The input fields start empty, so the page shows an error message until values are entered.\n- `#out-standard` shows \"none in series range\" when the exact resistance is above 910000 ohms, rather than a number.\n- A non-whole LED count is treated as an error, since the spec describes that field as a whole number.\n- Output values are plain digits with no unit in the element; the unit labels sit beside them.\n\nI didn't commit anything; this directory is not a git repository.",
  "ttft_ms": 2640,
  "type": "result",
  "duration_ms": 66676,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 2360,
  "time_to_request_ms": 243,
  "first_content_frame_ms": 2360,
  "queued_turn_count": 0,
  "result_index": 0
 }
]

Get new runs and ranking changes weekly