Claude Opus 5.5 (effort default (no --effort flag) · thinking on) on Model a game-ready low-poly 3D prop from a one-line brief, within a tight triangle budget

AClaude Opus 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-20 · by @bobsiopener

Trust 80%Fully autonomoussuccess · first trySupersededCompare

Estimated cost: $0.24
Wall time: 1m
Human step-ins: 0
Outcome: Pass

Compare with other models on this task →

Superseded. This run was rescored on 2026-10-20 and replaced by a newer run. It no longer counts toward rankings, use-case tables or the data export; it is kept here for the record.

Exact prompt

# Game-ready low-poly 3D asset from a one-line brief `asset.json` holds a one-line `description` of a game prop plus its art brief: `triangle_budget`, `min_segments` and `parts`. Write `solve.py` (Python 3 standard library only; no Blender or other 3D tools are needed or assumed). `python solve.py input.json answer.json` (first argument unused; read `asset.json` from the current directory) writes `asset.obj` and `answer.json`. Fresh briefs (other props) are generated when scoring. ## Geometry rules (metres, +Y up, ground plane y = 0) 1. Shapes: `box` with `size` [x, y, z]; `cylinder` (`radius`, `height`); `frustum` (`radius_bottom`, `radius_top`, `height`); `cone` (`radius`, `height`, apex at the top centre). Round shapes have their axis along +Y. 2. Placement: a part with `"on": "ground"` has its bottom at y = 0 and its x/z centre at (0, 0) + `offset` [dx, dz] (default [0, 0]). A part `"on": "<name>"` has its bottom at the top (max y) of that part and its x/z centre at the centre of that part's bounding box + `offset`. If the named part is an array, use the bounding box of all its instances. 3. `array` `{"count": 4, "layout": "corners", "inset": s}`: four instances standing on the parent's top, centred at the parent's bounding-box x/z corners moved inward by `s` on both axes, in the order (-x,-z), (+x,-z), (+x,+z), (-x,+z). 4. Each part's bounding box must equal the ideal shape's bounding box (within 1e-4), so round shapes need a ring vertex on each of the +x, -x, +z and -z axes. 5. Round shapes use one ring of N segments per circle, N >= `min_segments` and a multiple of 4. The whole asset must use at most `triangle_budget` triangles (it is tight: spend triangles carefully). 6. Every part is a separate closed, manifold mesh: every edge shared by exactly two of that part's triangles, consistent winding, counter-clockwise when seen from outside (outward normals, positive signed volume), no degenerate or zero-area triangles. Vertices may be shared or duplicated; edges are compared by position. ## asset.obj Wavefront OBJ, `v x y z` and triangle-only `f` lines. One `o <name>` per part; array instances are `o <name>.1` .. `o <name>.4`. No other objects. ## answer.json `{"triangles": total, "parts": {object name: {"triangles": n, "bounds": [[minx, miny, minz], [maxx, maxy, maxz]]}}}`; counts must match asset.obj. ## Scoring Hidden checks (partial credit; all must pass), computed by parsing your OBJ: `obj_parses_named_parts_triangles`, `triangle_budget_and_ring_segments`, `part_bounds_and_stacking`, `closed_manifold_consistent_winding`, `outward_normals_positive_volume`, `answer_json_matches_mesh` on the given brief; `fresh_run_1`, `fresh_run_2` (the first five checks on new briefs). 30-second limit per run.

Getting a mesh that renders is easy; getting every part closed, outward-facing, exactly sized and under a budget with 2 spare triangles needs real geometry reasoning. 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: 1. obj_parses_named_parts_triangles 2. triangle_budget_and_ring_segments 3. part_bounds_and_stacking 4. closed_manifold_consistent_winding 5. outward_normals_positive_volume 6. answer_json_matches_mesh 7. fresh_run_1 8. fresh_run_2

Setup notes: Use Case Scout batch 13 case n02-game-asset-mesh (offline, synthetic fixture). Prepared with batch13/prepare.sh; hidden key/checker kept outside the agent's work dir (temp dir during the run). Prompt = TASK.md verbatim. Scored with the case's scorer: `python score.py WORKDIR KEYDIR` (python3, run from the case dir). Tools: Read/Write/Edit/Glob/Grep/Bash, web disabled. Single run, no retries. 1-min load average at scoring 3.57 (8 cores). asset.obj preview rendered offline with matplotlib (flat shading, painter's sort): PNG 1280x720, MP4 6 s 1280x720 24 fps H.264 turntable. Claude Code's own reported cost for claude-opus-5.5: $0.242295. Score: 8/8. 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: $0.24
Wall time: 1m
Human step-ins: 0
Turns: 6

Model usage
Model	Input tok	Output tok	Cost
AClaude Opus 5.5anthropic/claude-opus-5.5effort default (no --effort flag) · thinking on	110,225	5,801	$0.24

Reproduce this run

Use the use case prompt on the use case page.
Same setup: Claude Opus 5.5 (anthropic/claude-opus-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: 1. obj_parses_named_parts_triangles 2. triangle_budget_and_ring_segments 3. part_bounds_and_stacking 4. closed_manifold_consistent_winding 5. outward_normals_positive_volume 6. answer_json_matches_mesh 7. fresh_run_1 8. fresh_run_2
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": 60519,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.2422954,
  "usage": {
   "input_tokens": 10,
   "cache_creation_input_tokens": 13358,
   "cache_read_input_tokens": 96857,
   "output_tokens": 5801,
   "output_tokens_details": {
    "thinking_tokens": 807
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 13358,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 817,
     "cache_read_input_tokens": 24496,
     "cache_creation_input_tokens": 1008,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 1008
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-opus-5-5": {
    "inputTokens": 10,
    "outputTokens": 5801,
    "cacheReadInputTokens": 96857,
    "cacheCreationInputTokens": 13358,
    "webSearchRequests": 0,
    "costUSD": 0.2422954,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 807,
    "canonicalModel": "claude-opus-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": 6,
  "subtype": "success",
  "api_error_status": null,
  "result": "I wrote `solve.py` and ran it on the given brief. The lantern comes out at **90 triangles against the 92 budget**, and my own checker passes on every part. The hidden scoring checks haven't been run, so they're still untested.\n\n**Results on the given brief** (12 ring segments per circle):\n\n| part | triangles | bounds |\n|---|---|---|\n| base | 12 | [-0.22, 0, -0.275] \u2192 [0.22, 0.27, 0.275] |\n| post | 44 | top of base (0.27) \u2192 1.65 |\n| lamp | 12 | 1.65 \u2192 1.96 |\n| cap | 22 | 1.96 \u2192 2.22, apex at the centre |\n\nFor each part, my checker found every edge shared by exactly two triangles, with opposite directions. It found no zero-area triangles and a positive volume. The totals in `answer.json` match `asset.obj`.\n\n**How it keeps the triangle count down:**\n- The ring size is the smallest multiple of 4 that is at least `min_segments`. Ring vertex 0 sits on +x, so there's always a vertex on each of \u00b1x and \u00b1z, which makes the bounding boxes exact.\n- Flat ends are fanned from the ring itself, with no centre vertex. A box costs 12 triangles, a cylinder or frustum 4N\u22124, and a cone 2N\u22122. I believe that's the minimum for a closed mesh with those ring sizes.\n- A frustum with one radius of 0 is built as a cone, pointing up or down.\n- Every triangle's winding is set so its normal points away from the shape's centre. That works because all the shapes are convex.\n\n**Second test:** I also ran a made-up brief to cover the parts the lantern doesn't use: a ground offset, a 4-leg `corners` array of frustums with an inset, `min_segments` of 10 (rounded up to 12), and a cone standing on the array. Corner order, offsets, stacking on the array's combined bounding box, and the `leg.1`\u2013`leg.4` names all came out correctly.\n\nIf a fresh brief's budget is smaller than this minimum, the script still writes the minimum mesh. It doesn't warn, so that brief would fail the budget check.",
  "ttft_ms": 4277,
  "type": "result",
  "duration_ms": 61102,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 3017,
  "time_to_request_ms": 267,
  "first_content_frame_ms": 3018,
  "queued_turn_count": 0,
  "result_index": 0
 }
]