{
  "run_id": "20261017T091535_deepseek-deepseek-v4.1-flash-1017_8287",
  "model": "deepseek/deepseek-v4.1-flash-1017",
  "provider": "DeepSeek",
  "effort": "high",
  "timestamp": "2026-10-17T09:15:35.327782+00:00",
  "task_count": 8,
  "failed_count": 0,
  "total_time_seconds": 5416.82,
  "time_complete": true,
  "usage_complete": false,
  "previous_attempts": 8,
  "previous_usage": [
    {
      "task_id": "aero-lab",
      "task_name": "Aero Lab",
      "task_number": 1,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 1.2977667079831008,
      "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": ""
    },
    {
      "task_id": "jet-canyon",
      "task_name": "Jet Odyssey",
      "task_number": 2,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.7391735000128392,
      "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": ""
    },
    {
      "task_id": "blast-island",
      "task_name": "Blast Island",
      "task_number": 3,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.7309253339772113,
      "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": ""
    },
    {
      "task_id": "launch-day",
      "task_name": "Launch Day",
      "task_number": 4,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.7734412919962779,
      "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": ""
    },
    {
      "task_id": "camera-shop",
      "task_name": "Alex Optics",
      "task_number": 5,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.7535918749927077,
      "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": ""
    },
    {
      "task_id": "afterburner",
      "task_name": "Afterburner",
      "task_number": 6,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.8258390830014832,
      "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": ""
    },
    {
      "task_id": "hangar-planner",
      "task_name": "Hangar Planner",
      "task_number": 7,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.8289332920103334,
      "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": ""
    },
    {
      "task_id": "rotor-rampage",
      "task_name": "Rotor Rampage",
      "task_number": 8,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 0.7897949579928536,
      "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": ""
    }
  ],
  "prompt_tokens": 29327616,
  "output_tokens": 970589,
  "output_tokens_basis": "exact",
  "reasoning_tokens": 510769,
  "tokens_per_second": 179.4,
  "cost_usd": 0.709398,
  "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": 2,
      "score": 7.4,
      "time_seconds": 1064.6111780409992,
      "ttft_ms": 273634.0763750195,
      "prompt_tokens": 7252669,
      "output_tokens": 143118,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 94296,
      "tokens_per_second": 134.43217857561174,
      "cost_usd": 0.114373902,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.0,
      "visual": 8.0,
      "code": 7.0,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.36,
      "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": 2,
      "score": 7.76,
      "time_seconds": 748.6737642500084,
      "ttft_ms": 79542.09795800853,
      "prompt_tokens": 2783183,
      "output_tokens": 123694,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 51970,
      "tokens_per_second": 165.2174897886421,
      "cost_usd": 0.08568220199999999,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.5,
      "visual": 8.3,
      "code": 6.8,
      "ux": 6.8,
      "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": 2,
      "score": 7.15,
      "time_seconds": 577.2488937500166,
      "ttft_ms": 108540.04700001678,
      "prompt_tokens": 3115749,
      "output_tokens": 129665,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 68482,
      "tokens_per_second": 224.62580942797393,
      "cost_usd": 0.097020678,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 8.0,
      "code": 7.0,
      "ux": 5.5,
      "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": 2,
      "score": 6.85,
      "time_seconds": 1113.4114243750228,
      "ttft_ms": 176763.6350840039,
      "prompt_tokens": 7552908,
      "output_tokens": 210634,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 140935,
      "tokens_per_second": 189.1789462446306,
      "cost_usd": 0.15353791199999997,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 7.0,
      "code": 6.5,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.36,
      "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": 2,
      "score": 8.0,
      "time_seconds": 734.6145699159824,
      "ttft_ms": 116686.99037499027,
      "prompt_tokens": 3936708,
      "output_tokens": 135041,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 77383,
      "tokens_per_second": 183.82564889156035,
      "cost_usd": 0.09888465599999999,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.5,
      "visual": 8.5,
      "code": 7.5,
      "ux": 7.0,
      "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": false,
      "attempts": 2,
      "score": 7.9,
      "time_seconds": 602.8471407920006,
      "ttft_ms": 95688.04737500614,
      "prompt_tokens": 1916507,
      "output_tokens": 104220,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 31790,
      "tokens_per_second": 172.8796455152449,
      "cost_usd": 0.070383474,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.5,
      "visual": 8.5,
      "code": 7.5,
      "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": "e08d0617005c33f9",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "06_afterburner.html",
      "screenshot": "06_afterburner_screenshot.jpg",
      "frames": "06_afterburner_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "hangar-planner",
      "task_name": "Hangar Planner",
      "task_number": 7,
      "wildcard": false,
      "failed": false,
      "attempts": 2,
      "score": 7.45,
      "time_seconds": 102.19207950000418,
      "ttft_ms": 70874.63279202348,
      "prompt_tokens": 166899,
      "output_tokens": 23474,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 8321,
      "tokens_per_second": 229.7046905675213,
      "cost_usd": 0.017142162,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.0,
      "visual": 7.5,
      "code": 8.0,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.97,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "14deca81c8b3a6c9",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "07_hangar-planner.html",
      "screenshot": "07_hangar-planner_screenshot.jpg",
      "frames": "07_hangar-planner_frames.jpg",
      "checks": [
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "reported",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "failed"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "hidden",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        },
        {
          "kind": "feature",
          "outcome": "first-try"
        }
      ],
      "lines_changed": 183,
      "repair": null
    },
    {
      "task_id": "rotor-rampage",
      "task_name": "Rotor Rampage",
      "task_number": 8,
      "wildcard": false,
      "failed": false,
      "attempts": 2,
      "score": 7.51,
      "time_seconds": 466.48611233301926,
      "ttft_ms": 71625.2580830187,
      "prompt_tokens": 2602993,
      "output_tokens": 100743,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 37592,
      "tokens_per_second": 215.96141307649623,
      "cost_usd": 0.07237327799999999,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.3,
      "visual": 8.2,
      "code": 7.0,
      "ux": 5.8,
      "innovation": null,
      "required_pass_rate": 0.5,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "d196c18e5127f818",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "08_rotor-rampage.html",
      "screenshot": "08_rotor-rampage_screenshot.jpg",
      "frames": "08_rotor-rampage_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    }
  ],
  "score": 7.5,
  "functional": 8.1,
  "visual": 8.0,
  "code": 7.16,
  "ux": 6.2,
  "innovation": 0.0,
  "required_pass_rate": 0.55,
  "wildcard_average": null,
  "ranked": true,
  "folder": "20261017t091535_deepseek-deepseek-v4.1-flash-1017_8287",
  "ran_with": "pi-openrouter",
  "evidence": {
    "kind": "benchmark",
    "model_details": {
      "display_name": "deepseek/deepseek-v4.1-flash-1017",
      "parameters_b": null,
      "active_parameters_b": null,
      "quantization": "",
      "hardware": "",
      "location": "cloud"
    },
    "official_score": 7.4319999999999995,
    "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": "86dba3a67d891fa1",
    "track_label": "pi-openrouter · 2 attempts",
    "coverage": {
      "scored": 5,
      "total": 5
    },
    "evidence_status": "measured",
    "attempts": {
      "aero-lab": 2,
      "blast-island": 2,
      "camera-shop": 2,
      "jet-canyon": 2,
      "launch-day": 2
    },
    "execution_harness": "pi-openrouter",
    "scoring_metadata": "recorded",
    "scoring_label": "claude-opus-5-5",
    "core_scores": {
      "functional": 8.0,
      "visual": 7.959999999999999,
      "code": 6.959999999999999,
      "ux": 6.26,
      "innovation": null
    }
  },
  "experiment": null,
  "run_average": 7.5,
  "official_score": 7.4319999999999995,
  "quality": {
    "standard": "quality-1",
    "label": "Quality v1",
    "score": 7.5,
    "coverage": {
      "scored": 8,
      "total": 8
    },
    "dimensions": {
      "functional": 8.1,
      "visual": 8.0,
      "code": 7.16,
      "ux": 6.2
    },
    "scorers": [
      "claude-code/claude-opus-5-5"
    ]
  }
}