{
  "aggregates": {
    "case": {
      "image-chart": {
        "attempts": 2,
        "latency_p50_seconds": 0.0016835,
        "latency_p95_seconds": 0.00181,
        "pass_rate": 0.0,
        "passed": 0
      },
      "image-ocr": {
        "attempts": 2,
        "latency_p50_seconds": 0.001813,
        "latency_p95_seconds": 0.001823,
        "pass_rate": 0.0,
        "passed": 0
      },
      "image-scene": {
        "attempts": 2,
        "latency_p50_seconds": 0.002003,
        "latency_p95_seconds": 0.002099,
        "pass_rate": 0.0,
        "passed": 0
      },
      "image-spatial-count": {
        "attempts": 2,
        "latency_p50_seconds": 0.001584,
        "latency_p95_seconds": 0.001626,
        "pass_rate": 0.0,
        "passed": 0
      },
      "image-ui": {
        "attempts": 2,
        "latency_p50_seconds": 0.0018419999999999999,
        "latency_p95_seconds": 0.001928,
        "pass_rate": 0.0,
        "passed": 0
      },
      "multi-image-comparison": {
        "attempts": 2,
        "latency_p50_seconds": 0.001633,
        "latency_p95_seconds": 0.001736,
        "pass_rate": 0.0,
        "passed": 0
      }
    },
    "modality": {
      "image": {
        "attempts": 12,
        "latency_p50_seconds": 0.0017794999999999998,
        "latency_p95_seconds": 0.002099,
        "pass_rate": 0.0,
        "passed": 0
      }
    }
  },
  "attempts": [
    {
      "case_id": "image-chart",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 12635,
          "kind": "image",
          "mime": "image/png",
          "path": "chart.png",
          "sha256": "c905e453c4e16eb274e46992d1a901bbced8f2314b47b9d408971bb176ab2f3e"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.00181,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-chart",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 12635,
          "kind": "image",
          "mime": "image/png",
          "path": "chart.png",
          "sha256": "c905e453c4e16eb274e46992d1a901bbced8f2314b47b9d408971bb176ab2f3e"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.001557,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-ocr",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 15916,
          "kind": "image",
          "mime": "image/png",
          "path": "ocr.png",
          "sha256": "a8d5158019fb2b08ee9f6a153860ab2342384644f7c53ab58fdd45658124e83e"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.001823,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-ocr",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 15916,
          "kind": "image",
          "mime": "image/png",
          "path": "ocr.png",
          "sha256": "a8d5158019fb2b08ee9f6a153860ab2342384644f7c53ab58fdd45658124e83e"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.001803,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-scene",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 13187,
          "kind": "image",
          "mime": "image/png",
          "path": "scene.png",
          "sha256": "30a1f11e40bf4e0a8595d39828e0aa362e8b28ac683c404a1068cc20bf7f7bda"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.002099,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-scene",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 13187,
          "kind": "image",
          "mime": "image/png",
          "path": "scene.png",
          "sha256": "30a1f11e40bf4e0a8595d39828e0aa362e8b28ac683c404a1068cc20bf7f7bda"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.001907,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-spatial-count",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 8785,
          "kind": "image",
          "mime": "image/png",
          "path": "spatial.png",
          "sha256": "d48f47cfba9d65524d6a9ca37d593550e10400c1b5fa160bec8aea24e0164051"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.001542,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-spatial-count",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 8785,
          "kind": "image",
          "mime": "image/png",
          "path": "spatial.png",
          "sha256": "d48f47cfba9d65524d6a9ca37d593550e10400c1b5fa160bec8aea24e0164051"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.001626,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-ui",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 16940,
          "kind": "image",
          "mime": "image/png",
          "path": "ui.png",
          "sha256": "03fd14d8b68973c915de2848736d22eb6f67ba6bdb5da2a7a563085ac894e804"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.001756,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "image-ui",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "media": [
        {
          "bytes": 16940,
          "kind": "image",
          "mime": "image/png",
          "path": "ui.png",
          "sha256": "03fd14d8b68973c915de2848736d22eb6f67ba6bdb5da2a7a563085ac894e804"
        }
      ],
      "modality": "image",
      "output": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.001928,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "multi-image-comparison",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "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": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 1,
      "sampling": null,
      "seconds": 0.001736,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    },
    {
      "case_id": "multi-image-comparison",
      "failure": {
        "message": "HTTP 400: {\"error\":{\"code\":\"vision_disabled\",\"message\":\"Vision is disabled for this server\",\"param\":\"messages\",\"type\":\"invalid_request_error\"}}",
        "type": "RuntimeError"
      },
      "finish_reason": null,
      "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": "",
      "passed": false,
      "reasoning": {
        "chars": 0,
        "field": null,
        "tokens": null
      },
      "repetition": 2,
      "sampling": null,
      "seconds": 0.00153,
      "usage": null,
      "validation": {
        "assertions": [],
        "finish_reason_allowed": false,
        "visible_answer": false
      }
    }
  ],
  "configuration": {
    "allowed_finish_reasons": [
      "stop"
    ],
    "chat_template_kwargs": null,
    "concurrency": 1,
    "max_images_per_request": 4,
    "max_tokens": 1024,
    "max_videos_per_request": 1,
    "reasoning_effort": null,
    "thinking_mode": "unsupported",
    "timeout_seconds": 120.0
  },
  "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": "Describe the scene and report every prominent label. Use the labels verbatim.",
        "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": "Transcribe all visible text exactly. Preserve every number.",
        "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": "Read the chart and report each quarter with its value in ascending quarter order.",
        "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": "Report the application name, status, and GPU percentage.",
        "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": "Count the blue boxes. Reply with the count written as an English word.",
        "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": "Compare the two meters. Identify which meter is higher and report its percentage.",
        "repetitions": 2,
        "sampling": null
      }
    ],
    "path": "tests\\fixtures\\eval-data\\agents-a1-v1\\image-corpus.json",
    "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": 12,
  "finished_at": "2026-09-19T18:07:58Z",
  "identity": {
    "base_url": "http://127.0.0.1:8081/v1",
    "engine_build_ref": null,
    "engine_revision": "a99407c63fc5bbd25d9fb597cbb8ab352bdb01ef",
    "hardware": "rtx5090",
    "model": "qwen38-huihui-ninfer-nvfp4-nospec-8k",
    "model_revision": "181446902fc777c479749e98cf2abf2250263a8d",
    "runtime_image": "nvidia/cuda:13.1.2-devel-ubuntu24.04@sha256:b9f64abf7226fdb3463ca202bc99878ec847171e6c5f77bd34c8d1403fbf1eca",
    "served_models": [
      "qwen38-huihui-ninfer-nvfp4-nospec-8k"
    ]
  },
  "passed": false,
  "passed_attempts": 0,
  "requests": 12,
  "run_id": "multimodal-20260919T180758Z",
  "schema": "multimodal-benchmark-evidence/v1",
  "started_at": "2026-09-19T18:07:58Z"
}