{
  "aggregates": {
    "case": {
      "image-chart": {
        "attempts": 3,
        "latency_p50_seconds": 0.522429,
        "latency_p95_seconds": 0.527442,
        "pass_rate": 1.0,
        "passed": 3
      },
      "image-ocr": {
        "attempts": 3,
        "latency_p50_seconds": 0.182364,
        "latency_p95_seconds": 0.187829,
        "pass_rate": 1.0,
        "passed": 3
      },
      "image-scene": {
        "attempts": 3,
        "latency_p50_seconds": 0.984736,
        "latency_p95_seconds": 1.17607,
        "pass_rate": 1.0,
        "passed": 3
      },
      "image-spatial-count": {
        "attempts": 3,
        "latency_p50_seconds": 0.120427,
        "latency_p95_seconds": 0.137985,
        "pass_rate": 1.0,
        "passed": 3
      },
      "image-ui": {
        "attempts": 3,
        "latency_p50_seconds": 0.275727,
        "latency_p95_seconds": 0.280495,
        "pass_rate": 1.0,
        "passed": 3
      },
      "multi-image-comparison": {
        "attempts": 3,
        "latency_p50_seconds": 0.533343,
        "latency_p95_seconds": 0.552479,
        "pass_rate": 1.0,
        "passed": 3
      }
    },
    "modality": {
      "image": {
        "attempts": 18,
        "latency_p50_seconds": 0.3995915,
        "latency_p95_seconds": 1.17607,
        "pass_rate": 1.0,
        "passed": 18
      }
    }
  },
  "attempts": [
    {
      "case_id": "image-chart",
      "failure": null,
      "finish_reason": "stop",
      "media": [
        {
          "bytes": 12635,
          "kind": "image",
          "mime": "image/png",
          "path": "chart.png",
          "sha256": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.522429,
      "usage": {
        "completion_tokens": 77,
        "prompt_tokens": 249,
        "total_tokens": 326
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.518688,
      "usage": {
        "completion_tokens": 77,
        "prompt_tokens": 249,
        "total_tokens": 326
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.527442,
      "usage": {
        "completion_tokens": 77,
        "prompt_tokens": 249,
        "total_tokens": 326
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.187829,
      "usage": {
        "completion_tokens": 21,
        "prompt_tokens": 245,
        "total_tokens": 266
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.181846,
      "usage": {
        "completion_tokens": 21,
        "prompt_tokens": 245,
        "total_tokens": 266
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.182364,
      "usage": {
        "completion_tokens": 21,
        "prompt_tokens": 245,
        "total_tokens": 266
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 1.17607,
      "usage": {
        "completion_tokens": 128,
        "prompt_tokens": 249,
        "total_tokens": 377
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.984736,
      "usage": {
        "completion_tokens": 128,
        "prompt_tokens": 249,
        "total_tokens": 377
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.967271,
      "usage": {
        "completion_tokens": 128,
        "prompt_tokens": 249,
        "total_tokens": 377
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.137985,
      "usage": {
        "completion_tokens": 2,
        "prompt_tokens": 249,
        "total_tokens": 251
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.120427,
      "usage": {
        "completion_tokens": 2,
        "prompt_tokens": 249,
        "total_tokens": 251
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.107527,
      "usage": {
        "completion_tokens": 2,
        "prompt_tokens": 249,
        "total_tokens": 251
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.280495,
      "usage": {
        "completion_tokens": 39,
        "prompt_tokens": 245,
        "total_tokens": 284
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.275725,
      "usage": {
        "completion_tokens": 39,
        "prompt_tokens": 245,
        "total_tokens": 284
      },
      "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": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.275727,
      "usage": {
        "completion_tokens": 39,
        "prompt_tokens": 245,
        "total_tokens": 284
      },
      "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": "[private fingerprint withheld]"
        },
        {
          "bytes": 7679,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-b.png",
          "sha256": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.552479,
      "usage": {
        "completion_tokens": 81,
        "prompt_tokens": 471,
        "total_tokens": 552
      },
      "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": "[private fingerprint withheld]"
        },
        {
          "bytes": 7679,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-b.png",
          "sha256": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.532918,
      "usage": {
        "completion_tokens": 81,
        "prompt_tokens": 471,
        "total_tokens": 552
      },
      "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": "[private fingerprint withheld]"
        },
        {
          "bytes": 7679,
          "kind": "image",
          "mime": "image/png",
          "path": "compare-b.png",
          "sha256": "[private fingerprint withheld]"
        }
      ],
      "modality": "image",
      "output": "[text withheld]",
      "passed": true,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 3,
      "sampling": null,
      "seconds": 0.533343,
      "usage": {
        "completion_tokens": 81,
        "prompt_tokens": 471,
        "total_tokens": 552
      },
      "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": 2048,
    "max_videos_per_request": 1,
    "reasoning_effort": null,
    "sampling": {
      "temperature": {
        "effective_request": 0.0,
        "requested": 0.0,
        "sent": true
      },
      "top_p": {
        "effective_request": null,
        "requested": null,
        "sent": false
      }
    },
    "thinking_mode": "unsupported",
    "timeout_seconds": 20.0
  },
  "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": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "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": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "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": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "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": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "sampling": null
      },
      {
        "assertions": [
          {
            "type": "contains_casefold",
            "value": "three"
          }
        ],
        "id": "image-spatial-count",
        "media": [
          {
            "bytes": 8785,
            "kind": "image",
            "mime": "image/png",
            "path": "spatial.png",
            "sha256": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "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": "[private fingerprint withheld]"
          },
          {
            "bytes": 7679,
            "kind": "image",
            "mime": "image/png",
            "path": "compare-b.png",
            "sha256": "[private fingerprint withheld]"
          }
        ],
        "modality": "image",
        "prompt": "[text withheld]",
        "repetitions": 3,
        "sampling": null
      }
    ],
    "path": "[private operator or asset text withheld]",
    "provenance": {
      "license": "CC0-1.0",
      "observed_date": "2026-08-15",
      "source": "[text withheld]"
    },
    "schema": "multimodal-corpus/v1",
    "sha256": "[private fingerprint withheld]"
  },
  "failed_attempts": 0,
  "finished_at": "2026-11-03T15:10:51Z",
  "identity": {
    "base_url": "[private operator or asset text withheld]",
    "engine_build_ref": null,
    "engine_revision": "[private fingerprint withheld]",
    "hardware": "RTX5090",
    "model": "[private identity withheld]",
    "model_revision": "[private identity withheld]",
    "runtime_image": "anvil-ninfer@[private fingerprint withheld]",
    "served_models": [
      "[private operator or asset text withheld]"
    ]
  },
  "passed": true,
  "passed_attempts": 18,
  "requests": 18,
  "run_id": "multimodal-20261103T151043Z",
  "schema": "multimodal-benchmark-evidence/v1",
  "started_at": "2026-11-03T15:10:43Z"
}