Reason-Constraint-Rate-Limit-Budget

The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.

Return JSON only with keys answer, evidence, constraints. A service allows 120 requests per minute and each job needs 3 requests. What is the maximum steady-state jobs per minute without exceeding the limit?

{
  "metrics": {
    "accuracy": {
      "checks": [
        {
          "contains": [
            "40"
          ]
        },
        {
          "contains": [
            "120"
          ]
        },
        {
          "contains": [
            "3"
          ]
        }
      ]
    },
    "evidence": {
      "checks": [
        {
          "contains": [
            "per minute"
          ]
        },
        {
          "contains": [
            "jobs"
          ]
        },
        {
          "contains": [
            "requests"
          ]
        }
      ]
    },
    "constraint": {
      "checks": [
        {
          "contains": [
            "constraint"
          ]
        },
        {
          "contains": [
            "throughput"
          ]
        },
        {
          "contains": [
            "not burst"
          ]
        }
      ]
    }
  }
}