{
  "run_id": "20261012T183411_meta-muse-wonder_a28a",
  "model": "meta/muse-wonder",
  "provider": "Meta",
  "effort": "max",
  "timestamp": "2026-10-12T18:34:11.192258+00:00",
  "task_count": 6,
  "failed_count": 1,
  "total_time_seconds": 3445.35,
  "time_complete": false,
  "usage_complete": false,
  "previous_attempts": 0,
  "previous_usage": [],
  "prompt_tokens": 2468868,
  "output_tokens": 228172,
  "output_tokens_basis": "exact",
  "reasoning_tokens": 106041,
  "tokens_per_second": 66.23,
  "cost_usd": 10.838883,
  "cost_basis": "api-equivalent-estimated",
  "cost_complete": false,
  "allowance": null,
  "tasks": [
    {
      "task_id": "aero-lab",
      "task_name": "Aero Lab",
      "task_number": 1,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.35,
      "time_seconds": 425.41880737501197,
      "ttft_ms": 228620.318375004,
      "prompt_tokens": 229614,
      "output_tokens": 32574,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 12900,
      "tokens_per_second": 76.56925231160646,
      "cost_usd": 1.328332,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.0,
      "visual": 7.5,
      "code": 7.0,
      "ux": 6.5,
      "innovation": null,
      "required_pass_rate": 0.45,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "1e8e809e8caeb2ba",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "01_aero-lab.html",
      "screenshot": "01_aero-lab_screenshot.jpg",
      "frames": "01_aero-lab_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "jet-canyon",
      "task_name": "Jet Odyssey",
      "task_number": 2,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 6.45,
      "time_seconds": 513.1778323750477,
      "ttft_ms": 117731.24045802979,
      "prompt_tokens": 725682,
      "output_tokens": 34713,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 13880,
      "tokens_per_second": 67.64321802316388,
      "cost_usd": 2.0594016,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 7.0,
      "code": 4.5,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.54,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "1ce617b6d67089b2",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "02_jet-canyon.html",
      "screenshot": "02_jet-canyon_screenshot.jpg",
      "frames": "02_jet-canyon_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "blast-island",
      "task_name": "Blast Island",
      "task_number": 3,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 6.5,
      "time_seconds": 398.7084040409536,
      "ttft_ms": 215529.97737500118,
      "prompt_tokens": 477220,
      "output_tokens": 43023,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 16372,
      "tokens_per_second": 107.90592714865589,
      "cost_usd": 2.0147208,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 7.5,
      "code": 5.0,
      "ux": 5.0,
      "innovation": null,
      "required_pass_rate": 0.6,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "28ba22cd0f56ca1e",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "03_blast-island.html",
      "screenshot": "03_blast-island_screenshot.jpg",
      "frames": "03_blast-island_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "launch-day",
      "task_name": "Launch Day",
      "task_number": 4,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 6.75,
      "time_seconds": 1476.1085550409625,
      "ttft_ms": 991937.4186249916,
      "prompt_tokens": 594006,
      "output_tokens": 78051,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 46439,
      "tokens_per_second": 52.87619242734763,
      "cost_usd": 3.4090416,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 7.0,
      "code": 6.0,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.45,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "d1e76d48a7ac8eb6",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "04_launch-day.html",
      "screenshot": "04_launch-day_screenshot.jpg",
      "frames": "04_launch-day_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "camera-shop",
      "task_name": "Alex Optics",
      "task_number": 5,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.25,
      "time_seconds": 631.9381737079821,
      "ttft_ms": 270282.46279101586,
      "prompt_tokens": 442346,
      "output_tokens": 39811,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 16450,
      "tokens_per_second": 62.99825150046501,
      "cost_usd": 2.0273872,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.0,
      "visual": 7.5,
      "code": 7.0,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.6,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "de7c304e41dd3505",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "05_camera-shop.html",
      "screenshot": "05_camera-shop_screenshot.jpg",
      "frames": "05_camera-shop_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "afterburner",
      "task_name": "Afterburner",
      "task_number": 6,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": null,
      "ttft_ms": null,
      "prompt_tokens": null,
      "output_tokens": null,
      "output_tokens_basis": "unavailable",
      "reasoning_tokens": null,
      "tokens_per_second": null,
      "cost_usd": null,
      "cost_basis": "",
      "functional": 0.0,
      "visual": 0.0,
      "code": 0.0,
      "ux": 0.0,
      "innovation": 0.0,
      "required_pass_rate": 0.0,
      "judge": "error",
      "scored_by": "",
      "prompt_fingerprint": "e08d0617005c33f9",
      "scoring_fingerprint": "a1fabb77c780d380",
      "rubric": "",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "interrupted",
      "continuations": 0,
      "html": null,
      "screenshot": null,
      "frames": null,
      "checks": [],
      "lines_changed": null,
      "repair": null
    }
  ],
  "score": 6.86,
  "functional": 7.7,
  "visual": 7.3,
  "code": 5.9,
  "ux": 5.9,
  "innovation": 0.0,
  "required_pass_rate": 0.44,
  "wildcard_average": null,
  "ranked": true,
  "folder": "20261012t183411_meta-muse-wonder_a28a",
  "ran_with": "pi-openrouter",
  "evidence": {
    "kind": "benchmark",
    "model_details": {
      "display_name": "meta/muse-wonder",
      "parameters_b": null,
      "active_parameters_b": null,
      "quantization": "",
      "hardware": "",
      "location": "cloud"
    },
    "official_score": 6.859999999999999,
    "edition": {
      "aggregation": "equal-weight-mean-v1",
      "generation": {
        "max_continuations": 2,
        "max_html_size_bytes": 5242880,
        "max_tokens": 262144,
        "recording_duration_seconds": 15,
        "sandbox_timeout_seconds": 30,
        "screenshot_height": 1080,
        "screenshot_width": 1920,
        "temperature": null
      },
      "id": "e4c5f05fc9fb2725",
      "name": "VisualForge Core",
      "tasks": [
        {
          "id": "aero-lab",
          "name": "Aero Lab",
          "number": 1,
          "prompt_fingerprint": "1e8e809e8caeb2ba",
          "rubric_fingerprint": "fed23c28cf8bdd2d",
          "weight": 1
        },
        {
          "id": "blast-island",
          "name": "Blast Island",
          "number": 3,
          "prompt_fingerprint": "28ba22cd0f56ca1e",
          "rubric_fingerprint": "d8234e0468a8a9be",
          "weight": 1
        },
        {
          "id": "camera-shop",
          "name": "Alex Optics",
          "number": 5,
          "prompt_fingerprint": "de7c304e41dd3505",
          "rubric_fingerprint": "5542170a0e786535",
          "weight": 1
        },
        {
          "id": "jet-canyon",
          "name": "Jet Odyssey",
          "number": 2,
          "prompt_fingerprint": "1ce617b6d67089b2",
          "rubric_fingerprint": "2db92ee795650b26",
          "weight": 1
        },
        {
          "id": "launch-day",
          "name": "Launch Day",
          "number": 4,
          "prompt_fingerprint": "d1e76d48a7ac8eb6",
          "rubric_fingerprint": "02187e0e817733ad",
          "weight": 1
        }
      ]
    },
    "scoring_revision": "60735ef367139bc4",
    "track": "f13f477d940269b4",
    "track_label": "pi-openrouter · 1 attempt",
    "coverage": {
      "scored": 5,
      "total": 5
    },
    "evidence_status": "measured",
    "attempts": {
      "aero-lab": 1,
      "blast-island": 1,
      "camera-shop": 1,
      "jet-canyon": 1,
      "launch-day": 1
    },
    "execution_harness": "pi-openrouter",
    "scoring_metadata": "recorded",
    "scoring_label": "claude-opus-5-5",
    "core_scores": {
      "functional": 7.7,
      "visual": 7.3,
      "code": 5.9,
      "ux": 5.9,
      "innovation": null
    }
  },
  "experiment": null,
  "run_average": 6.86,
  "official_score": 6.859999999999999,
  "quality": {
    "standard": "quality-1",
    "label": "Quality v1",
    "score": 6.86,
    "coverage": {
      "scored": 5,
      "total": 6
    },
    "dimensions": {
      "functional": 7.7,
      "visual": 7.3,
      "code": 5.9,
      "ux": 5.9
    },
    "scorers": [
      "claude-code/claude-opus-5-5"
    ]
  }
}
