{
  "campaign_record_sha256": "b55ccee2f73023cf554fcad4745cadba3ebf50274edddb1346842f6163753699",
  "checks": [
    {
      "case_id": "gpt120",
      "case_record_sha256": "ccc093e5f70b89e37e1922afd2f1b64fd4417d0a342bee809502be03568e70b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "answer": 391
      },
      "failure_type": "none",
      "outer_fence_present": false,
      "primary_exact_expected_answer": true,
      "primary_expected_final_answer_valid": true,
      "primary_strict_json_object_valid": true,
      "response_record_sha256": "697fab42317414c176566dcec485964d0aff3e8464fdea0eeba3626ae2c8d675",
      "response_stream_sha256": "303a7e304734760650fd170b96ea125b14cdbdb0f38a9074ce982441f3dca714",
      "task": "arithmetic",
      "visible_final_answer": "{\n  \"answer\": 391\n}"
    },
    {
      "case_id": "gpt120",
      "case_record_sha256": "ccc093e5f70b89e37e1922afd2f1b64fd4417d0a342bee809502be03568e70b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "company": "Northstar Labs",
        "region": "EU"
      },
      "failure_type": "none",
      "outer_fence_present": false,
      "primary_exact_expected_answer": true,
      "primary_expected_final_answer_valid": true,
      "primary_strict_json_object_valid": true,
      "response_record_sha256": "88e634fe1163f39aa3e6301f8d5ad23776d67e6f4cb4038ed98a9a4209ed3f56",
      "response_stream_sha256": "db1454704feed345bcea621a3aa5ba8c7111261ae8aca73a8c19b1652cc7267b",
      "task": "extraction",
      "visible_final_answer": "{\n  \"company\": \"Northstar Labs\",\n  \"region\": \"EU\"\n}"
    },
    {
      "case_id": "gpt120",
      "case_record_sha256": "ccc093e5f70b89e37e1922afd2f1b64fd4417d0a342bee809502be03568e70b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "category": "large"
      },
      "failure_type": "none",
      "outer_fence_present": false,
      "primary_exact_expected_answer": true,
      "primary_expected_final_answer_valid": true,
      "primary_strict_json_object_valid": true,
      "response_record_sha256": "aafe5b6bf5cf1481047325fe3fe70bc4e6b30b4d6d18f7193fb3429b4ab99fa8",
      "response_stream_sha256": "0f41c68c6df80b20b73da8ef2141d7718adb054414d075091440830a27a4c4da",
      "task": "classification",
      "visible_final_answer": "{\n  \"category\": \"large\"\n}"
    },
    {
      "case_id": "qwen72-q8",
      "case_record_sha256": "81835d12b4546153652824e253a63fe3ea0ae9d559d4d09f2b97a5c16b7026b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "answer": 391
      },
      "failure_type": "none",
      "outer_fence_present": false,
      "primary_exact_expected_answer": true,
      "primary_expected_final_answer_valid": true,
      "primary_strict_json_object_valid": true,
      "response_record_sha256": "e45914b4d04183e66ad3cc84f512d4b6d4984e342aecb80ee959c8cae949ab4e",
      "response_stream_sha256": "be11284c9954c39f8e323675538ae469bb0e184bbd045177f24fd32fffa9555b",
      "task": "arithmetic",
      "visible_final_answer": "{\"answer\": 391}"
    },
    {
      "case_id": "qwen72-q8",
      "case_record_sha256": "81835d12b4546153652824e253a63fe3ea0ae9d559d4d09f2b97a5c16b7026b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "company": "Northstar Labs",
        "region": "EU"
      },
      "failure_type": "single_outer_markdown_json_fence_only",
      "outer_fence_present": true,
      "primary_exact_expected_answer": false,
      "primary_expected_final_answer_valid": false,
      "primary_strict_json_object_valid": false,
      "response_record_sha256": "0a0373c9bc4171e97bb8813fadc55462e2060b2d0663b3d244e71a0610e4997a",
      "response_stream_sha256": "e13c57c31e81999d5e92453d3093a6cde344a36dd1f681d57e2166a839f6d36d",
      "task": "extraction",
      "visible_final_answer": "```json\n{\n  \"company\": \"Northstar Labs\",\n  \"region\": \"EU\"\n}\n```"
    },
    {
      "case_id": "qwen72-q8",
      "case_record_sha256": "81835d12b4546153652824e253a63fe3ea0ae9d559d4d09f2b97a5c16b7026b7",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "category": "large"
      },
      "failure_type": "single_outer_markdown_json_fence_only",
      "outer_fence_present": true,
      "primary_exact_expected_answer": false,
      "primary_expected_final_answer_valid": false,
      "primary_strict_json_object_valid": false,
      "response_record_sha256": "53980606b6dd3b256c2e64a25b35d555035dfa09b7fc29f6b017558d8b3b0392",
      "response_stream_sha256": "9db732ce5dec2bf729d3f5094b09941ee8fa849415a679a041b36afc9c8eabae",
      "task": "classification",
      "visible_final_answer": "```json\n{\n  \"category\": \"large\"\n}\n```"
    },
    {
      "case_id": "qwen235-hybrid",
      "case_record_sha256": "b4596ba0b10a895ef31922bf705fe96f452ddcf91a94761ce31ee2d94bdc656c",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "answer": 391
      },
      "failure_type": "single_outer_markdown_json_fence_only",
      "outer_fence_present": true,
      "primary_exact_expected_answer": false,
      "primary_expected_final_answer_valid": false,
      "primary_strict_json_object_valid": false,
      "response_record_sha256": "ec0fe6f36bdb777305f9411753c9bd6d3f3b114a31f25e938882316872b02ec2",
      "response_stream_sha256": "e7c57a40264fdbafb0b39a72695ec99b6cece144d9b2f3fe90a8a3df3f5da707",
      "task": "arithmetic",
      "visible_final_answer": "```json\n{\"answer\": 391}\n```"
    },
    {
      "case_id": "qwen235-hybrid",
      "case_record_sha256": "b4596ba0b10a895ef31922bf705fe96f452ddcf91a94761ce31ee2d94bdc656c",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "company": "Northstar Labs",
        "region": "EU"
      },
      "failure_type": "none",
      "outer_fence_present": false,
      "primary_exact_expected_answer": true,
      "primary_expected_final_answer_valid": true,
      "primary_strict_json_object_valid": true,
      "response_record_sha256": "0f49d758c63404ac645851d9f37bedccb57b2e9fe149e2db1ad5ad367af3c476",
      "response_stream_sha256": "1b0fd7b8623e6a6ee1dc7fa75c5545df6fb589b9711efd818aeed6cedcc802d8",
      "task": "extraction",
      "visible_final_answer": "{\n  \"company\": \"Northstar Labs\",\n  \"region\": \"EU\"\n}"
    },
    {
      "case_id": "qwen235-hybrid",
      "case_record_sha256": "b4596ba0b10a895ef31922bf705fe96f452ddcf91a94761ce31ee2d94bdc656c",
      "exact_after_single_outer_fence_removal": true,
      "expected_object": {
        "category": "large"
      },
      "failure_type": "single_outer_markdown_json_fence_only",
      "outer_fence_present": true,
      "primary_exact_expected_answer": false,
      "primary_expected_final_answer_valid": false,
      "primary_strict_json_object_valid": false,
      "response_record_sha256": "7559d4c7009ef4989434f719f716cce8d683e8038c7622e78c0edf6dda9a2763",
      "response_stream_sha256": "46797923ff40a94030cc918108b42e7567cd93e74a17ac38dc3414f15bab41d8",
      "task": "classification",
      "visible_final_answer": "```json\n{\n  \"category\": \"large\"\n}\n```"
    }
  ],
  "format_rule": {
    "arbitrary_prefix_suffix_stripping": false,
    "complete_outer_fence_only": true,
    "fence_language": "json",
    "maximum_fence_removals": 1,
    "reject_nested_or_multiple_fences": true,
    "unfenced_answer": "unchanged"
  },
  "historical_results_pooled": false,
  "interpretation": {
    "en": "The original strict-JSON result remains 5/9. All 9 answers contained the expected values; 4 had an additional Markdown fence. This reviews the format of the same answers, not a retest or a general assessment of model quality.",
    "pl": "Pierwotny wynik ścisłego JSON pozostaje 5/9. Wszystkie 9 odpowiedzi zawierało oczekiwane wartości; 4 miały dodatkową ramkę Markdown. To przegląd formatu tych samych odpowiedzi, nie ponowny test ani ogólna ocena jakości modeli."
  },
  "kind": "capacity_practical_format_review",
  "model_results": [
    {
      "case_id": "gpt120",
      "checks": 3,
      "primary_strict_valid": 3,
      "values_match_after_format_review": 3
    },
    {
      "case_id": "qwen72-q8",
      "checks": 3,
      "primary_strict_valid": 1,
      "values_match_after_format_review": 3
    },
    {
      "case_id": "qwen235-hybrid",
      "checks": 3,
      "primary_strict_valid": 1,
      "values_match_after_format_review": 3
    }
  ],
  "outer_fence_count": 4,
  "plan_sha256": "4c4a17de3e1f76e9cf66873b3268855214359059c5ccacaaf48e6af97fee3e4c",
  "presentation_only_review": true,
  "primary_exporter_sha256": "358debee510861b5cfbc9937ca8bed3812eb36bd41d3228ada8174664e417a68",
  "primary_strict_results_unchanged": true,
  "primary_strict_valid_count": 5,
  "retest": false,
  "reviewed_public_sha256": "e4419081851d69f351673aee406bafe401b6ebe71f15c1499f2c753e6127540f",
  "runner_sha256": "c01ffbddca9cfc65f5dab27529c93166816a73c0cdbefb53d1b9f95adc844166",
  "schema": 1,
  "scope": "presentation-only review, not a retest; primary strict results unchanged",
  "source_review_sha256": "fefeeb0f4ab58dbe6334a70f41f801985a46ea8228a7892d2225ea555efe4772",
  "total_checks": 9,
  "values_match_after_format_review_count": 9
}
