Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) on Animate an online order's parcel moving along its delivery route with depot stops in one self-contained HTML file

AClaude Haiku 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-24 · by @bobsiopener

Trust 80%Fully autonomoussuccess · first tryCompare

Estimated cost: $0.01
Wall time: 48s
Human step-ins: 0
Outcome: Pass

Compare with other models on this task →

Exact prompt

# Animate an online order's parcel moving along its delivery route with depot stops in one self-contained HTML file Make a small animated scene: a parcel delivery tracker. Write one file, `out/scene.html`, that works opened straight from disk with no network access. `spec.json` gives the page title, the stage size and the scene parameters. **Looks are not scored**: the page is opened in a headless browser (800x600 window, UTC time zone, network blocked) and checked by objective tests only. ## Scene contract - Draw the scene on one `<canvas>` or `<svg>` stage of `width` x `height` CSS pixels (spec.json), visible on load, with `<title>` containing spec.json `title`. - Animate continuously from load (`requestAnimationFrame` or timers); no click is needed unless stated below. - Expose `window.sceneState()`, returning the scene's current state in stage pixel coordinates (origin top-left of the stage): `{parcel: {x, y}, status: "in_transit" | "at <stop name>" | "delivered"} (canvas px, y down)`. Extra fields are fine. It must describe what is actually drawn. - Time 0 is `performance.now()` 0 (seconds = `performance.now()` / 1000): the parcel leaves the first stop at time 0. Leg i runs from stops[i] to stops[i + 1] at leg_speeds_px_s[i]; after legs that end at an intermediate stop it waits dwell_s[i] seconds there before leaving. It stays at the last stop once it arrives. Draw the route, the stops and the parcel. `sceneState()` must return the state at the moment it is called. - The page must log no errors and throw no exceptions. Keep the file at most 200,000 bytes; no external scripts, fonts, images or URLs (inline everything). ## Checks (partial credit = share passed) 1. `self_contained`: `out/scene.html` exists, is at most 200,000 bytes, references no http(s) or protocol-relative URL, and requests nothing from the network when opened. 2. `has_canvas_or_svg`: a visible `<canvas>` or `<svg>` of at least 300x200 CSS pixels is on the page. 3. `frames_change`: over 3 s, some later screenshot differs from the first in at least 0.1% of pixels, and at least 3 of 30 screenshots differ. 4. `route_position`: at every sample, the parcel is within 1.5 px of its exact position on the route: straight legs between consecutive `stops` at each leg's `leg_speeds_px_s`, waiting `dwell_s` at each intermediate stop. 5. `tracking_status`: at every sample more than 0.05 s from an arrival or departure, `status` is in_transit while moving, at <stop name> while waiting at an intermediate stop, and delivered after reaching the last stop; at least one sample falls inside a stop. 6. `no_console_errors`: no uncaught exception, `console.error` or browser error is logged during the run. ## 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.

Make a parcel delivery tracker animation as one self-contained HTML file. A headless browser checks objective facts only: the file is offline and small, a canvas or SVG stage is drawn, frames change, the scene follows a timed specification (route position; tracking status) and nothing errors. A poster PNG and a short MP4 are captured for the gallery. 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 has_canvas_or_svg frames_change route_position tracking_status no_console_errors

Setup notes: Batch 34 case tq07-parcel-tracking-animation (offline, synthetic inputs, seed 20809246). 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 0.13 (8 cores). Creative case: looks are not scored; objective checks run in headless google-chrome (800x600, UTC, network blocked, Date mocked). Gallery media (3 s capture: poster.png + clip.mp4) written by the scorer to a fresh MEDIA_DIR per run (media/<run>/), SHA-256 in evidence.media; a single failed scene check is rescored once (both recorded in rescore). 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: $0.010767. Score: 6/6. 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.

Output

Metrics

Estimated cost: $0.01
Wall time: 48s
Human step-ins: 0
Turns: 7

Model usage
Model	Input tok	Output tok	Cost
AClaude Haiku 5.5anthropic/claude-haiku-5.5effort default (no --effort flag) · thinking on	185,448	10,155	$0.01

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: self_contained has_canvas_or_svg frames_change route_position tracking_status no_console_errors
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": 45670,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.010766869999999998,
  "usage": {
   "input_tokens": 14,
   "cache_creation_input_tokens": 20177,
   "cache_read_input_tokens": 165257,
   "output_tokens": 10155,
   "output_tokens_details": {
    "thinking_tokens": 4556
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 20177,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 686,
     "cache_read_input_tokens": 31925,
     "cache_creation_input_tokens": 791,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 791
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-haiku-5-5": {
    "inputTokens": 14,
    "outputTokens": 10155,
    "cacheReadInputTokens": 165257,
    "cacheCreationInputTokens": 20177,
    "webSearchRequests": 0,
    "costUSD": 0.010766869999999998,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 4556,
    "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": 7,
  "subtype": "success",
  "api_error_status": null,
  "result": "I wrote `out/scene.html`, a self-contained 5.9 KB page that draws the route with its four stops and animates the parcel along it. I checked it in headless Chrome and against a separate timeline calculation in Node, but I did not run the six graded checks themselves.\n\n- **Timeline:** The parcel leaves the Harbour sort centre at t=0, arrives at Eastfield depot at 2.52 s, waits 0.8 s, reaches Hilltop depot at 4.17 s, waits 0.6 s, and arrives at the Customer door at 6.07 s. It stays there and reports `delivered` from then on.\n- **`window.sceneState()`:** Returns `{parcel: {x, y}, status, time_s, leg}` computed from `performance.now()/1000`. The same function drives the drawing, so the reported state matches what is on screen.\n- **Verification:** In Node, I compared 1,601 samples over 8 s against an independent calculation. The largest position error was 0.0000 px, and there were no status mismatches outside the 0.05 s windows around stop arrivals and departures. In headless Chrome, the page rendered the route, stops and parcel with no page errors. The only console output was dbus noise from the sandbox.\n- **Choices I made:**\n  - The spec values are copied into the page as constants, because a `file://` page can't reliably fetch `spec.json`.\n  - The page has no network references and no favicon request (`<link rel=\"icon\" href=\"data:,\">`).\n  - The stage is 720x480 CSS px, as spec.json says, and sits at the top of the 800x600 window.\n  - A pulse ring draws around the parcel while it waits at a depot.",
  "ttft_ms": 1331,
  "type": "result",
  "duration_ms": 46915,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 1137,
  "time_to_request_ms": 154,
  "first_content_frame_ms": 1137,
  "queued_turn_count": 0,
  "result_index": 0
 }
]