{
  "run_id": "20261008T144524_stepfun-step-5-preview_e97b",
  "model": "stepfun/step-5-preview",
  "provider": "StepFun",
  "effort": "high",
  "timestamp": "2026-10-08T14:45:24.747278+00:00",
  "task_count": 9,
  "failed_count": 2,
  "total_time_seconds": 18964.45,
  "time_complete": true,
  "usage_complete": false,
  "previous_attempts": 0,
  "previous_usage": [],
  "prompt_tokens": 54100294,
  "output_tokens": 1069903,
  "output_tokens_basis": "exact",
  "reasoning_tokens": 0,
  "tokens_per_second": 77.36,
  "cost_usd": 6.240123,
  "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": 5.8,
      "time_seconds": 1309.917814415996,
      "ttft_ms": 405428.3624160016,
      "prompt_tokens": 8783447,
      "output_tokens": 212450,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 162.18574758044429,
      "cost_usd": 1.0956188,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 4.5,
      "code": 6.0,
      "ux": 5.0,
      "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": 7.2,
      "time_seconds": 1424.4048213339993,
      "ttft_ms": 216408.02625000651,
      "prompt_tokens": 16066813,
      "output_tokens": 173597,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 121.87335889345069,
      "cost_usd": 1.4197328999999999,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.0,
      "visual": 7.0,
      "code": 7.0,
      "ux": 6.5,
      "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": 6.4,
      "time_seconds": 1686.0649620839977,
      "ttft_ms": 879261.6881669965,
      "prompt_tokens": 8375598,
      "output_tokens": 139206,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 82.56265513514948,
      "cost_usd": 0.8753006000000001,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.0,
      "visual": 7.0,
      "code": 6.0,
      "ux": 5.0,
      "innovation": null,
      "required_pass_rate": 0.5,
      "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": 4.65,
      "time_seconds": 2478.3132790829986,
      "ttft_ms": 836005.8223749948,
      "prompt_tokens": 12515976,
      "output_tokens": 258343,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 104.24146219948011,
      "cost_usd": 1.4907757000000001,
      "cost_basis": "api-equivalent-estimated",
      "functional": 5.0,
      "visual": 3.5,
      "code": 6.0,
      "ux": 4.5,
      "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": null,
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "camera-shop",
      "task_name": "Alex Optics",
      "task_number": 5,
      "wildcard": false,
      "failed": true,
      "attempts": 1,
      "score": 0.0,
      "time_seconds": 3638.970909833006,
      "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": "de7c304e41dd3505",
      "scoring_fingerprint": "fccdeb941276edac",
      "rubric": "",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "error",
      "continuations": 0,
      "html": null,
      "screenshot": null,
      "frames": null,
      "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": 1495.8445450419968,
      "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": "error",
      "continuations": 0,
      "html": null,
      "screenshot": null,
      "frames": null,
      "checks": [],
      "lines_changed": null,
      "repair": null
    },
    {
      "task_id": "hangar-planner",
      "task_name": "Hangar Planner",
      "task_number": 7,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.2,
      "time_seconds": 384.2340508339985,
      "ttft_ms": 257005.85854200472,
      "prompt_tokens": 480261,
      "output_tokens": 40594,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 105.64914773141207,
      "cost_usd": 0.181532,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.5,
      "visual": 7.5,
      "code": 7.5,
      "ux": 6.0,
      "innovation": null,
      "required_pass_rate": 0.94,
      "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": "failed"
        },
        {
          "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": "after-feedback"
        },
        {
          "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": 197,
      "repair": null
    },
    {
      "task_id": "rotor-rampage",
      "task_name": "Rotor Rampage",
      "task_number": 8,
      "wildcard": false,
      "failed": false,
      "attempts": 1,
      "score": 7.7,
      "time_seconds": 4372.51538225,
      "ttft_ms": 1703826.5213749981,
      "prompt_tokens": 6949329,
      "output_tokens": 212711,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 48.64728455009885,
      "cost_usd": 0.9847303000000001,
      "cost_basis": "api-equivalent-estimated",
      "functional": 8.5,
      "visual": 8.5,
      "code": 7.0,
      "ux": 6.0,
      "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
    },
    {
      "task_id": "wild-cursive-ink-grand-prix",
      "task_name": "Cursive Ink Grand Prix",
      "task_number": 9,
      "wildcard": true,
      "failed": false,
      "attempts": 1,
      "score": 6.4,
      "time_seconds": 2174.1889291660045,
      "ttft_ms": 1597552.2093749896,
      "prompt_tokens": 928870,
      "output_tokens": 33002,
      "output_tokens_basis": "exact",
      "reasoning_tokens": 0,
      "tokens_per_second": 15.178993673129966,
      "cost_usd": 0.192433,
      "cost_basis": "api-equivalent-estimated",
      "functional": 7.0,
      "visual": 7.0,
      "code": 6.0,
      "ux": 5.0,
      "innovation": null,
      "required_pass_rate": 0.5,
      "judge": "llm-vision",
      "scored_by": "claude-code/claude-opus-5-5 (quality-1 rubric)",
      "prompt_fingerprint": "e1846c1d94dd5759",
      "scoring_fingerprint": "59929116d9ab5b1e",
      "rubric": "quality-1",
      "evaluation_history": [],
      "score_note": "",
      "finish_reason": "stop",
      "continuations": 0,
      "html": "09_wild-cursive-ink-grand-prix.html",
      "screenshot": "09_wild-cursive-ink-grand-prix_screenshot.jpg",
      "frames": "09_wild-cursive-ink-grand-prix_frames.jpg",
      "checks": [],
      "lines_changed": null,
      "repair": null
    }
  ],
  "score": 6.49,
  "functional": 7.25,
  "visual": 6.33,
  "code": 6.58,
  "ux": 5.5,
  "innovation": 0.0,
  "required_pass_rate": 0.41,
  "wildcard_average": 6.4,
  "ranked": true,
  "folder": "20261008t144524_stepfun-step-5-preview_e97b",
  "ran_with": "pi-openrouter",
  "evidence": {
    "kind": "benchmark",
    "model_details": {
      "display_name": "stepfun/step-5-preview",
      "parameters_b": null,
      "active_parameters_b": null,
      "quantization": "",
      "hardware": "",
      "location": "cloud"
    },
    "official_score": null,
    "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": "",
    "track": "f13f477d940269b4",
    "track_label": "pi-openrouter · 1 attempt",
    "coverage": {
      "scored": 4,
      "total": 5
    },
    "evidence_status": "incomplete",
    "attempts": {
      "aero-lab": 1,
      "blast-island": 1,
      "camera-shop": 1,
      "jet-canyon": 1,
      "launch-day": 1
    },
    "execution_harness": "pi-openrouter",
    "scoring_metadata": "recorded"
  },
  "experiment": null,
  "run_average": 6.49,
  "official_score": null,
  "quality": {
    "standard": "quality-1",
    "label": "Quality v1",
    "score": 6.49,
    "coverage": {
      "scored": 6,
      "total": 8
    },
    "dimensions": {
      "functional": 7.25,
      "visual": 6.33,
      "code": 6.58,
      "ux": 5.5
    },
    "scorers": [
      "claude-code/claude-opus-5-5"
    ]
  }
}