{
  "schema_version": "remakebench.public-evidence.v1",
  "snapshot": {
    "benchmark_id": "tech-review-001",
    "benchmark_version": "v1",
    "immutable": true
  },
  "result_id": "jungle-temple-fable-5-medium",
  "task": {
    "id": "jungle-temple-blender-mcp",
    "title": "Jungle Temple voxel diorama",
    "benchmark_prompt_version": "v1"
  },
  "configuration": {
    "id": "fable-medium",
    "model": "claude-fable-5",
    "reasoning": "Medium",
    "provider_agent": "Anthropic Claude Code"
  },
  "status": "partial_token_timing_and_artifact_ledger",
  "measurements": {
    "wall_clock_seconds": 724.2,
    "wall_clock_display": "12:04.2",
    "total_processed_tokens": 2519008,
    "reported_cost_usd": 10.1
  },
  "cost_provenance": {
    "currency": "USD",
    "reported_total_basis": "per-ledger rounded total",
    "calculation_location": "api_cost.primary_5m_cache_write_calculation.total_usd",
    "calculated_total_usd": 10.1047085,
    "estimate_type": null,
    "pricing_basis": "Anthropic first-party Claude API, standard global pricing",
    "pricing_mode": null,
    "cache_write_assumption": "5-minute cache writes",
    "verified_live_pricing": true,
    "pricing_accessed_on": "2026-07-12",
    "pricing_sources": [
      "https://platform.claude.com/docs/en/about-claude/pricing"
    ]
  },
  "primary_artifact": {
    "kind": "blender-scene",
    "sha256": "d9448b31f645dc0555697d65a7861d07832a55ca47aaa0d9c8071525bd98fcdf"
  },
  "public_standardized_captures": [
    {
      "kind": "poster",
      "url": "/harness/blender/jungle-temple/b3m8/poster.webp"
    },
    {
      "kind": "video",
      "url": "/harness/blender/jungle-temple/b3m8/turntable.mp4"
    }
  ],
  "caveats": [
    "Wall-clock is end-to-end latency including tool execution and user think-time, not model-only compute.",
    "Output tokens include hidden reasoning, code, and tool-call JSON, not just visible text.",
    "Cache-read tokens are billed at a deep discount, so total processed tokens greatly overstate effective cost.",
    "The primary total uses the supplied five-minute cache-write rate; a one-hour cache write would increase the cost.",
    "Blender and render environment, final capture metadata, and blind-evaluation evidence have not been supplied."
  ],
  "missing_evidence": [
    "Blender and render environment",
    "final capture metadata",
    "blind-evaluation record"
  ],
  "source_provenance": {
    "harness_commit": "3f14507a8ce536d68a1f472f34f4c1a64c511f82",
    "source_record_sha256": "cb5dd9c3d36f70613bdd482a2f8935d1aab1cdcebbeb4a929ecad13dcc2ae94b",
    "primary_artifact_sha256_verified_at_source_commit": true
  }
}