{
  "benchmark": "Nordform Trueline",
  "note": "Every dated run, in order, losses included. A run is never removed or replaced.",
  "runs": [
    {
      "date": "2026-10-20",
      "label": "v0 shakedown",
      "corpus": "factory-corpus@0",
      "split": "all",
      "file": "results/2026-10-20-v0-shakedown.json",
      "factoryModel": "claude-opus-5-5",
      "modelAloneRun": false,
      "plannerToArtifact": {
        "all": {
          "verifiedArtifacts": "23/28",
          "typedProposal": "53/60",
          "falseClaims": "0/22",
          "overRefusals": "0/38"
        },
        "dev": {
          "verifiedArtifacts": "19/23",
          "typedProposal": "39/44",
          "falseClaims": "0/17",
          "overRefusals": "0/27"
        },
        "heldOut": {
          "verifiedArtifacts": "4/5",
          "typedProposal": "14/16",
          "falseClaims": "0/5",
          "overRefusals": "0/11"
        }
      }
    },
    {
      "date": "2026-10-21",
      "label": "nightly 1cc74897",
      "corpus": "plumb-v1",
      "split": "all",
      "file": "results/2026-10-21-nightly-1cc74897.json",
      "factoryModel": "claude-opus-5-5",
      "modelAloneRun": true,
      "plannerToArtifact": {
        "all": {
          "verifiedArtifacts": "24/28",
          "typedProposal": "51/60",
          "falseClaims": "0/22",
          "overRefusals": "0/38"
        },
        "dev": {
          "verifiedArtifacts": "20/23",
          "typedProposal": "39/44",
          "falseClaims": "0/17",
          "overRefusals": "0/27"
        },
        "heldOut": {
          "verifiedArtifacts": "4/5",
          "typedProposal": "12/16",
          "falseClaims": "0/5",
          "overRefusals": "0/11"
        }
      }
    },
    {
      "date": "2026-10-21",
      "label": "nightly c2dd1170",
      "corpus": "plumb-v1",
      "split": "all",
      "file": "results/2026-10-21-nightly-c2dd1170.json",
      "factoryModel": "claude-opus-5-5",
      "modelAloneRun": true,
      "plannerToArtifact": {
        "all": {
          "verifiedArtifacts": "24/28",
          "typedProposal": "51/60",
          "falseClaims": "0/22",
          "overRefusals": "0/38"
        },
        "dev": {
          "verifiedArtifacts": "20/23",
          "typedProposal": "38/44",
          "falseClaims": "0/17",
          "overRefusals": "0/27"
        },
        "heldOut": {
          "verifiedArtifacts": "4/5",
          "typedProposal": "13/16",
          "falseClaims": "0/5",
          "overRefusals": "0/11"
        }
      }
    }
  ]
}
