{
  "aggregates": {
    "case": {
      "image-chart": {
        "attempts": 2,
        "latency_p50_seconds": 0.8680685,
        "latency_p95_seconds": 1.071595,
        "pass_rate": 1.0,
        "passed": 2
      },
      "image-ocr": {
        "attempts": 2,
        "latency_p50_seconds": 0.5616625,
        "latency_p95_seconds": 0.793235,
        "pass_rate": 1.0,
        "passed": 2
      },
      "image-scene": {
        "attempts": 2,
        "latency_p50_seconds": 1.6972225,
        "latency_p95_seconds": 2.071888,
        "pass_rate": 1.0,
        "passed": 2
      },
      "image-spatial-count": {
        "attempts": 2,
        "latency_p50_seconds": 0.7684340000000001,
        "latency_p95_seconds": 0.888302,
        "pass_rate": 1.0,
        "passed": 2
      },
      "image-ui": {
        "attempts": 2,
        "latency_p50_seconds": 1.087728,
        "latency_p95_seconds": 1.154176,
        "pass_rate": 1.0,
        "passed": 2
      },
      "multi-image-comparison": {
        "attempts": 2,
        "latency_p50_seconds": 1.5873754999999998,
        "latency_p95_seconds": 1.6213,
        "pass_rate": 1.0,
        "passed": 2
      }
    },
    "modality": {
      "image": {
        "attempts": 12,
        "latency_p50_seconds": 1.0464375000000001,
        "latency_p95_seconds": 2.071888,
        "pass_rate": 1.0,
        "passed": 12
      }
    }
  },
  "attempts": [
    {
      "case_id": "image-chart",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 12635,
          "kind": "image",
          "mime": "image/png",
          "path": "chart.png",
          "sha256": "c905e453c4e16eb274e46992d1a901bbced8f2314b47b9d408971bb176ab2f3e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 438,
        "field": "reasoning_content",
        "tokens": 127
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.664542,
      "usage": {
        "completion_tokens": 149,
        "completion_tokens_details": {
          "reasoning_tokens": 127
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 438
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "ordered_casefold",
            "values": [
              "Q1",
              "10",
              "Q2",
              "20",
              "Q3",
              "30"
            ]
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-chart",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 12635,
          "kind": "image",
          "mime": "image/png",
          "path": "chart.png",
          "sha256": "c905e453c4e16eb274e46992d1a901bbced8f2314b47b9d408971bb176ab2f3e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 674,
        "field": "reasoning_content",
        "tokens": 191
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 1.071595,
      "usage": {
        "completion_tokens": 210,
        "completion_tokens_details": {
          "reasoning_tokens": 191
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 499
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "ordered_casefold",
            "values": [
              "Q1",
              "10",
              "Q2",
              "20",
              "Q3",
              "30"
            ]
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-ocr",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 15916,
          "kind": "image",
          "mime": "image/png",
          "path": "ocr.png",
          "sha256": "a8d5158019fb2b08ee9f6a153860ab2342384644f7c53ab58fdd45658124e83e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 167,
        "field": "reasoning_content",
        "tokens": 36
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.33009,
      "usage": {
        "completion_tokens": 58,
        "completion_tokens_details": {
          "reasoning_tokens": 36
        },
        "prompt_tokens": 285,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 343
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "ANVIL READY"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42917"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-ocr",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 15916,
          "kind": "image",
          "mime": "image/png",
          "path": "ocr.png",
          "sha256": "a8d5158019fb2b08ee9f6a153860ab2342384644f7c53ab58fdd45658124e83e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 288,
        "field": "reasoning_content",
        "tokens": 83
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.793235,
      "usage": {
        "completion_tokens": 105,
        "completion_tokens_details": {
          "reasoning_tokens": 83
        },
        "prompt_tokens": 285,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 390
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "ANVIL READY"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42917"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-scene",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 13187,
          "kind": "image",
          "mime": "image/png",
          "path": "scene.png",
          "sha256": "30a1f11e40bf4e0a8595d39828e0aa362e8b28ac683c404a1068cc20bf7f7bda"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 1326,
        "field": "reasoning_content",
        "tokens": 352
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 2.071888,
      "usage": {
        "completion_tokens": 482,
        "completion_tokens_details": {
          "reasoning_tokens": 352
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 771
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "red car"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "green park"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "sunny day"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-scene",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 13187,
          "kind": "image",
          "mime": "image/png",
          "path": "scene.png",
          "sha256": "30a1f11e40bf4e0a8595d39828e0aa362e8b28ac683c404a1068cc20bf7f7bda"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 741,
        "field": "reasoning_content",
        "tokens": 172
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 1.322557,
      "usage": {
        "completion_tokens": 310,
        "completion_tokens_details": {
          "reasoning_tokens": 172
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 284
        },
        "total_tokens": 599
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "red car"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "green park"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "sunny day"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-spatial-count",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 8785,
          "kind": "image",
          "mime": "image/png",
          "path": "spatial.png",
          "sha256": "d48f47cfba9d65524d6a9ca37d593550e10400c1b5fa160bec8aea24e0164051"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 650,
        "field": "reasoning_content",
        "tokens": 159
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.648566,
      "usage": {
        "completion_tokens": 162,
        "completion_tokens_details": {
          "reasoning_tokens": 159
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 451
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "three"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-spatial-count",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 8785,
          "kind": "image",
          "mime": "image/png",
          "path": "spatial.png",
          "sha256": "d48f47cfba9d65524d6a9ca37d593550e10400c1b5fa160bec8aea24e0164051"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 776,
        "field": "reasoning_content",
        "tokens": 183
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.888302,
      "usage": {
        "completion_tokens": 186,
        "completion_tokens_details": {
          "reasoning_tokens": 183
        },
        "prompt_tokens": 289,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 475
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "three"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-ui",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 16940,
          "kind": "image",
          "mime": "image/png",
          "path": "ui.png",
          "sha256": "03fd14d8b68973c915de2848736d22eb6f67ba6bdb5da2a7a563085ac894e804"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 765,
        "field": "reasoning_content",
        "tokens": 184
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 1.02128,
      "usage": {
        "completion_tokens": 210,
        "completion_tokens_details": {
          "reasoning_tokens": 184
        },
        "prompt_tokens": 285,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 495
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "ANVIL SERVING"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "READY"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "image-ui",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 16940,
          "kind": "image",
          "mime": "image/png",
          "path": "ui.png",
          "sha256": "03fd14d8b68973c915de2848736d22eb6f67ba6bdb5da2a7a563085ac894e804"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 647,
        "field": "reasoning_content",
        "tokens": 149
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 1.154176,
      "usage": {
        "completion_tokens": 174,
        "completion_tokens_details": {
          "reasoning_tokens": 149
        },
        "prompt_tokens": 285,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 459
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "ANVIL SERVING"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "READY"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "multi-image-comparison",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 8000,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-a.png",
          "sha256": "7dac4f31d62c88d765150aa6051efea9ddaaa8c64312080be4c5b8f6ad96322a"
        },
        {
          "bytes": 7679,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-b.png",
          "sha256": "8f7cdfd733f1ee9731de6160e77af6aa50189edec6eaf9ddd7654a0c8300a43e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 1146,
        "field": "reasoning_content",
        "tokens": 296
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 1.6213,
      "usage": {
        "completion_tokens": 322,
        "completion_tokens_details": {
          "reasoning_tokens": 296
        },
        "prompt_tokens": 511,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 833
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "meter b"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "75"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    },
    {
      "case_id": "multi-image-comparison",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 8000,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-a.png",
          "sha256": "7dac4f31d62c88d765150aa6051efea9ddaaa8c64312080be4c5b8f6ad96322a"
        },
        {
          "bytes": 7679,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-b.png",
          "sha256": "8f7cdfd733f1ee9731de6160e77af6aa50189edec6eaf9ddd7654a0c8300a43e"
        }
      ],
      "modality": "image",
      "output": "[redacted]",
      "passed": true,
      "reasoning": {
        "chars": 1148,
        "field": "reasoning_content",
        "tokens": 284
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 1.553451,
      "usage": {
        "completion_tokens": 311,
        "completion_tokens_details": {
          "reasoning_tokens": 284
        },
        "prompt_tokens": 511,
        "prompt_tokens_details": {
          "cached_tokens": 0
        },
        "total_tokens": 822
      },
      "validation": {
        "assertions": [
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "meter b"
          },
          {
            "passed": true,
            "type": "contains_casefold",
            "value": "75"
          }
        ],
        "finish_reason_allowed": true,
        "visible_answer": true
      }
    }
  ],
  "configuration": {
    "allowed_finish_reasons": [
      "stop"
    ],
    "chat_template_kwargs": null,
    "concurrency": 1,
    "max_images_per_request": 4,
    "max_tokens": 8192,
    "max_videos_per_request": 1,
    "reasoning_effort": null,
    "sampling": {
      "temperature": {
        "effective_request": 0.9,
        "requested": 0.9,
        "sent": true
      },
      "top_p": {
        "effective_request": 0.95,
        "requested": 0.95,
        "sent": true
      }
    },
    "thinking_mode": "default",
    "timeout_seconds": 300.0
  },
  "corpus": {
    "cases": [
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "red car"
          },
          {
            "type": "contains_casefold",
            "value": "green park"
          },
          {
            "type": "contains_casefold",
            "value": "sunny day"
          }
        ],
        "id": "image-scene",
        "media": [
          {
            "bytes": 13187,
            "kind": "image",
            "mime": "image/png",
            "path": "scene.png",
            "sha256": "30a1f11e40bf4e0a8595d39828e0aa362e8b28ac683c404a1068cc20bf7f7bda"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "ANVIL READY"
          },
          {
            "type": "contains_casefold",
            "value": "42917"
          },
          {
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "id": "image-ocr",
        "media": [
          {
            "bytes": 15916,
            "kind": "image",
            "mime": "image/png",
            "path": "ocr.png",
            "sha256": "a8d5158019fb2b08ee9f6a153860ab2342384644f7c53ab58fdd45658124e83e"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "ordered_casefold",
            "values": [
              "Q1",
              "10",
              "Q2",
              "20",
              "Q3",
              "30"
            ]
          }
        ],
        "id": "image-chart",
        "media": [
          {
            "bytes": 12635,
            "kind": "image",
            "mime": "image/png",
            "path": "chart.png",
            "sha256": "c905e453c4e16eb274e46992d1a901bbced8f2314b47b9d408971bb176ab2f3e"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "ANVIL SERVING"
          },
          {
            "type": "contains_casefold",
            "value": "READY"
          },
          {
            "type": "contains_casefold",
            "value": "42"
          }
        ],
        "id": "image-ui",
        "media": [
          {
            "bytes": 16940,
            "kind": "image",
            "mime": "image/png",
            "path": "ui.png",
            "sha256": "03fd14d8b68973c915de2848736d22eb6f67ba6bdb5da2a7a563085ac894e804"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "three"
          }
        ],
        "id": "image-spatial-count",
        "media": [
          {
            "bytes": 8785,
            "kind": "image",
            "mime": "image/png",
            "path": "spatial.png",
            "sha256": "d48f47cfba9d65524d6a9ca37d593550e10400c1b5fa160bec8aea24e0164051"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "meter b"
          },
          {
            "type": "contains_casefold",
            "value": "75"
          }
        ],
        "id": "multi-image-comparison",
        "media": [
          {
            "bytes": 8000,
            "kind": "image",
            "mime": "image/png",
            "path": "compare-a.png",
            "sha256": "7dac4f31d62c88d765150aa6051efea9ddaaa8c64312080be4c5b8f6ad96322a"
          },
          {
            "bytes": 7679,
            "kind": "image",
            "mime": "image/png",
            "path": "compare-b.png",
            "sha256": "8f7cdfd733f1ee9731de6160e77af6aa50189edec6eaf9ddd7654a0c8300a43e"
          }
        ],
        "modality": "image",
        "prompt": "[redacted]",
        "repetitions": 2,
        "sampling": null
      }
    ],
    "path": "[operator identity redacted]",
    "provenance": {
      "derived_from": "corpus.json",
      "observed_date": "2026-08-29",
      "scope": "Deterministic image-only qualification subset for servers that intentionally disable video",
      "synthetic": {
        "generator": "generate.py",
        "license": "CC0-1.0"
      }
    },
    "schema": "multimodal-corpus/v1",
    "sha256": "a514ae146b5aab3bcb55ae9fb27f55a41791a20863304d1bb95947e93931cd12"
  },
  "failed_attempts": 0,
  "finished_at": "2026-10-05T08:51:44Z",
  "identity": {
    "base_url": "[operator identity redacted]",
    "engine_build_ref": null,
    "engine_revision": "bace20dc70249eed6402b66d4852c6c3f9612905",
    "hardware": "rtx5090",
    "model": "qwen38-swift15-ninfer-dflash2-c4-262k",
    "model_revision": "ff891a1130adfcf46bf745e06f0fa3dcde2e8c82",
    "runtime_image": "anvil-ninfer@sha256:59df48dd99f0d177b73de57ba88eb6457ed7e4b0c5771c45e8976adcd9f5ef67",
    "served_models": [
      "qwen38-swift15-ninfer-dflash2-c4-262k"
    ]
  },
  "passed": true,
  "passed_attempts": 12,
  "requests": 12,
  "run_id": "multimodal-20261005T085131Z",
  "schema": "multimodal-benchmark-evidence/v1",
  "started_at": "2026-10-05T08:51:31Z"
}