Measured harness ledgerPublic result
Claude Opus 5Courtyard Blossom Stage 2 — Claude Opus 5 Max
Create a second cherry-blossom temple courtyard arena for the fixed fighting game under the pinned reference-guided Stage 2 contract.
Max reasoningHeadline result
- Workflow cost
- $98.04
- Wall-clock
- 1h 20m 48.3s wall-clock
- Processed tokens
- 140.99M processed
- Record state
- artifact_and_final_metrics_ledger_no_independent_browser_replay
Public summary
Claude Opus 5 Max artifact_and_final_metrics_ledger_no_independent_browser_replay ledger: 1h 20m 48.3s wall-clock, 140.99M processed, and $98.04 Claude API list-price equivalent, not an itemized Claude Code subscription charge.
Run identity and stack
- Result ID: courtyard-arena-blossom-stage-2-opus-5-max
- Technical model: Claude Opus 5
- Provider: Anthropic Claude Code
- Stack: Anthropic Claude Code
- Stack: Technical model/configuration: Claude Opus 5
- Stack: Blender / browser arena workflow
- Stack: Reference-guided Stage 2 fixture
Cost basis
- All-5-minute cache-write alternative: $93.85.
Primary artifact integrity
- Kind: manifest-verified-artifact
- Path: artifacts/courtyard-arena-blossom-stage-2-opus-5-max/source/web/models/sakura.glb
- SHA-256: 3cdff30d7ad96c2755abf18053c431279c0659223c61abbc4129f52811b8afcb
Recorded caveats
- Wall-clock is end-to-end workflow latency, not model-only compute.
- Output tokens include hidden reasoning, generated code, and tool-call JSON.
- Cache reads are discounted, so total processed tokens substantially overstate cost.
- The archive contains the Stage 2 delta only. Reproduction requires the frozen Stage 1 starter and the same model's archived Stage 1 delta.
- The source receipt does not publicly establish user-turn count or operator-follow-up count.
- The only supplied game captures and FPS claims are model-produced. The archive operator confirmed the two HTTP entry points but could not run an independent browser replay because browser automation was unavailable.
- No blind evaluation, standardized sustained-play FPS/hitch receipt, or independent visual-quality score is recorded.
Public result only
This result keeps its public summary and evidence, but it does not currently have a matching Builder test with prompts, projects, Harness workflows or skills.